World of Warcraft: Forever launches November 4, 2026 — rates below are launch-day rates, nothing is on sale yet

WoW Forever Gold → Guides → Dungeon gold

Guide · income and the bill behind it

What a Dungeon Run Is Worth

The coin on the corpses pays for the repair afterwards and not much else. The money is in what you are allowed to sell, and that turns out to be the part players fight about.

Where the value actually sits

A levelling dungeon produces 4 kinds of value, and they are not equal:

  • Coin. Predictable, small, and mostly consumed by repairs and reagents on the way out.
  • Materials. Cloth above all, plus whatever the skinners in the group take off the corpses. Unglamorous and reliable.
  • Recipes and blueprints. The scarce end of crafting comes from bosses, which is new in Forever and makes a dungeon a supply line rather than a detour.
  • Bind-on-equip drops. The only items that can be sold to a stranger, and therefore the only ones with a real price.

Everything else you loot is yours forever whether you want it or not. The bind-on-equip drop is the one piece of a run that converts into gold at market rate, which is why it carries all the social weight.

It is worth saying why anyone queues at all, because the usual reason is weaker here. Experience from dungeon clears has been cut, so an instance is no longer the fastest way to move a character through its level band, and the old habit of chaining the same door until the bar fills does not pay the way it used to. What is left is the loot line: bind-on-equip drops, recipes and blueprints off bosses, and the cloth and skins that come off the trash on the way through. That is a materials and gold motive rather than a levelling one, and it changes how a run should be judged.

The practical difference is that a clear now has to justify itself against whatever else the same hour could produce, instead of being levelling progress with loot attached. A group that wipes twice and finds nothing sellable has lost the evening outright, where previously the experience would have covered it. We do not have figures for how far the experience was reduced and are not inventing any, but the direction is enough to plan around: pick doors for what they drop and for the crafting supply they feed, and count the result against the gold sinks of levelling the same way you would count a gathering route.

What exists instead of a published reduction is a handful of numbers testers wrote down during beta. They are player counts rather than anything official, they come from single runs, and they may not match what ships, but they are the only figures anyone has:

  • Deadmines: trash reported at 35 experience a mob against 50 for the same kill before the change.
  • The Stockades: 3950 experience total for a clear that took 32 minutes.
  • Blackfathom Deeps: 4929 experience for a run of 45 minutes.

Read those against the time attached to them and the conclusion is the same one the loot line already points at: an instance hour is no longer competitive with questing as a way of moving the bar, so there is little reason to judge it that way. The figure that decides whether the group forms again is gold and sellable materials per hour, with the approach and the assembly counted in. If a clear returns a thin bag and a repair bill, the experience will not quietly cover the difference any more, and the honest comparison is against whatever else that hour could have produced towards the running costs of levelling.

One official figure has since arrived. A hotfix on 8 November raised dungeon experience from 40% to 60% of the Classic Era value, and Blizzard described itself as satisfied with the levelling pace that produced. That is the only adjustment of its kind so far, and the wording suggests no further increase is being planned, so plans built on the lower number should be updated once rather than left open.

For the ledger the direction does not change. At 60% of Classic Era an instance hour still moves the bar more slowly than questing does, so the case for forming a group remains the loot line: bind-on-equip drops, boss recipes and blueprints, and the cloth and skins off the trash. The hotfix narrows the gap enough that a clear with the attached quests in the log is easier to justify, but a dungeon is still best counted as income first and progress second when it is weighed against a surface route or against the wider Forever gold economy.

The part that changes how a run is planned is where the missing experience went. Kill experience inside an instance was cut, and the bulk of what a dungeon still awards sits in the quests attached to it instead. A clear done without those quests picked up pays only the loot line and the materials line, because the mobs themselves no longer carry the difference. A clear done with the full set in the log pays the loot line plus a lump at the turn-in, and the 2 outcomes are far enough apart that they should not be described with the same word.

That makes the pre-run step worth more than the pulls. Collecting the quests means a trip to the givers before the group forms, and for some of them the chain starts outside the instance entirely, so a player who queues straight in has already given up most of the reward. The practical rule is simple: if the quests are not in the log, judge the hour purely as a gathering hour and compare it against a surface route on bags filled and gold earned, because that is all it will produce. If they are in the log, the run stands on its own and the materials become the bonus rather than the whole case for going. Either way the repair and reagent bill is still deducted first, which is where the standing costs of levelling decide whether the evening was worth assembling 5 people for.

There is a second cost that the loot line does not show, and it has grown since the experience cut. Fewer players queue for a door they no longer need, so groups form more slowly and at some hours do not form at all. The time spent in a channel asking for a tank is part of the dungeon hour whether or not a single pull happens, and it has to be counted against a gathering route that starts producing the moment the character lands. A clear that pays well on paper can still lose to a mediocre surface circuit once 40 minutes of assembly sit in front of it.

A third cost shows up once the group out-levels the door. Trash that has turned grey gives nothing back on the kill, so a party sitting above the instance band is paying repair and time for the loot line alone. The cloth and the skins still drop, and the boss recipes still drop, but the mobs between them have stopped contributing anything else. That is fine if the plan was always a materials run, and it is a poor plan if anyone in the group still expects the clear to move their bar.

One door that keeps coming up in these arguments is Blackmaw Hold, which is not in the Classic Era list and so has no established numbers behind it. The claim being passed around is that a clear returns roughly 100 g an hour from vendor trash alone, before anything sellable to another player is counted. That figure comes from a forum post by a single player describing their own runs, not from a parse, a log or anything published, and nobody has posted a second set of runs next to it. Treat it as hearsay with a number attached.

If the claim holds it changes what the instance is for rather than how it is run. Income that lands as grey and white items sold to a vendor is unaffected by what the auction house is doing that week, which makes it steadier than a bind-on-equip line but also caps it: there is no good night. It also means the bag fills quickly, so the realistic hour includes the walk out and the sale, not just the pulls. Until somebody records their own runs through the door, the honest way to use the number is as a reason to try Blackmaw Hold once and count the result yourself, measured against the repair bill and the rest of the standing costs of levelling.

That zeroing rule is what splits a dungeon trip into 2 separate activities, and they want different things. A run entered for money and loot is indifferent to grey trash, because none of its value was ever in the kill. It picks the door by what drops there, and out-levelling the instance helps rather than hurts: the pulls are faster, wipes are rare, the repair bill stays small, and the quests can be skipped entirely since nothing is being turned in. On that plan the only questions are bags filled, bind-on-equip drops seen, and how long the clear took against the wider Forever gold economy.

A run entered for experience has the opposite constraints. It has to happen while the mobs still count, which means staying inside the level band and going before the band is passed, and it has to have the attached quests collected first or most of the award is left behind. That narrows the choice of door to whatever is current for the group rather than whatever is richest, and it puts a deadline on the trip that a materials run does not have. The practical step is to settle which run it is before anyone pulls, because a group holding both intentions will argue about pace, about skipping bosses, and about whether a detour for cloth was worth the time.

The same threshold cuts the other way for the individual. A character that has pushed well past the band a door was built for is no longer a useful invite: the people still inside it are there for the bar, and an over-levelled member thins what the rest of the group takes from each kill. So the player who got ahead of the content finds the channel quiet, not because the dungeon stopped paying, but because nobody at the right level wants the seat filled that way. The loot line is still there and the access to it is not.

That leaves 2 practical routes, and both are worth pricing before the evening starts. The first is paying another player to run the instance, which turns the clear into a purchase: whatever the bind-on-equip drops, the boss recipes and the cloth return has to clear the fee first, and the fee is agreed in advance while the drops are not. The second is going in alone, which is slower per pull and only works on doors far enough below the character to be safe, but it removes the fee, removes the assembly wait and keeps the whole loot line in one bag with no rolls to argue about. Judged that way a solo clear is a gathering route with bosses attached, and it should be counted against a surface circuit on bags filled per hour and against the standing costs of levelling rather than treated as progress.

The other argument against farming a door repeatedly is the set pieces themselves. Players report them dropping rarely enough that no reliable schedule can be built on them, and we have no published rate to quote, so treat them as an occasional result rather than the reason to form a group. A run assembled to chase a specific piece will usually finish without it, after the same assembly wait and the same repair bill. Judged that way, a loot-motivated clear tends to come out behind a gathering circuit of the same length, and the comparison should be made before the group forms rather than after, with the repair and reagent line counted among the standing costs of levelling.

The change in motive also changes how the group treats a drop. When a clear was mainly experience, a bind-on-equip item was luck on top of progress that everyone was making anyway. Now the progress is thin and the item is the run, so players price it against what they are actually saving for, which at this stage is usually a mount and the ability training that arrives with it. A sellable drop is a visible slice of that fund rather than a bonus, and everyone at the table knows it, which is why need rolls on items nobody intends to wear produce arguments that the same roll would not have caused before. The honest way to read those disputes is as a disagreement about money, not about gear, and they are easier to settle if the group says out loud before the first pull whether bind-on-equip drops are rolled by everyone, reserved for the class that would wear them, or sold with the proceeds split.

The second effect is that runs are increasingly one-off. A group assembled to farm a door for a level band used to stay together for several clears; a group assembled to pull a specific drop or a specific recipe disbands once the attempt is made, because there is no bar filling in the background to justify a second lap. That makes the assembly time matter more per clear and makes a reliable set of names worth keeping, since re-forming from scratch is the expensive part. It also means the run should be judged the way a gathering trip is judged: bags filled, items sold, and the repair and reagent bill deducted first, with the remainder measured against the mount and training bills it is meant to cover.

There is also a level gap rule that decides where a paid run can still deliver anything at all. When the difference between group members is more than 12 levels, experience is zeroed for the run, so a high character dragging a much lower one through a door hands over loot and a fast clear and nothing on the bar. The threshold is not applied retroactively: dungeons that were already started keep running under the rules they began with, and only the next instance opened after the change is measured against the gap.

For anyone buying or selling a run for gold this sets the boundary of what is actually being sold. Inside 12 levels a carry still moves the character forward and the price can reflect that. Outside it the service is a loot service, and should be priced and described as one: bind-on-equip drops, boss recipes and blueprints, and whatever cloth and skins come off the trash. Sellers who keep quoting levelling speed across a wider gap are quoting something the run will not produce, and buyers who pay for it are paying twice, since the repair and reagent bill still comes out first and still has to be weighed against the wider Forever gold economy.

A later hotfix set the boundary for who may stand in the group. Experience is reduced to nothing only when the level difference inside the party is more than 12 levels, so a character being helped by someone moderately above them still earns from the run, while a pull led by somebody far outside that window pays no experience at all. The practical effect is that a dungeon stays a shared activity rather than a solo delivery: the helper has to be close enough in level to count, which is also the shape most paid runs take when one slot is filled by a stronger character and the rest of the group plays normally.

The hotfix did not reach instances that were already running when it went live. Groups inside at the time kept the old behaviour until the instance was reset, which is worth remembering if a run appears to contradict the rule: the state of the instance decides, not the state of the patch. For planning purposes the safe assumption is to form the group fresh, check the spread of levels before the first pull, and judge the evening on loot and materials against the standing costs of levelling as usual.

Krol'dok Stronghold is the clearest example of a door that has to be prepared for rather than queued into. It covers the 40-45 bracket and sits in Riverglades, and the quest set attached to it is not identical on both sides: Alliance characters can pick up 5 quests tied to the place, Horde characters 4. Since most of what a dungeon still awards sits in those turn-ins rather than in the kills, the faction difference is worth knowing before the group is assembled, and so is the walk to the givers.

The rest of the Krol'dok numbers fill in the gap the table had between the middle brackets. The instance holds 6 bosses, and the quest line around it comes to 10 in total: 9 split by faction, with Alliance taking 5 and Horde 4, plus 1 that both sides can pick up. There is also a set tied to the place whose bonus turns the wearer into an ogre, which matters to the ledger only because transformation gear tends to hold a price with buyers who are not chasing stats, so a spare piece is worth listing rather than vendoring.

The part worth planning for is the key. Access runs through a quest that consumes materials at the turn-in, so the cost of entry is paid in crafted or gathered goods before the first pull, and it is paid by whoever holds the key rather than split across the group. Work that out in advance: either the group agrees to cover the materials between them, or one player carries the bill and should be taking a larger share of what sells afterwards. Counted honestly, the key sits alongside repair and reagent costs as money spent before any income arrives, and a run that produces nothing sellable has still consumed it.

One of those quests, Key Ingredients, is a materials hand-in rather than a kill or escort objective: it asks for 12 Iron Bar and 4 Thick Leather. That is an entry fee paid before the first pull, per character who wants the reward, and it is either bought at market price or produced by a miner and a skinner in advance. Either way it belongs in the same column as repairs and reagents when the evening is weighed up, which is the column the standing costs of levelling already tracks. A group that walks in without the bars and the leather either skips the quest or pays auction prices on the spot, and the second option is usually the expensive one.

A rule change on 13 November tightened that further. Where a mob is grey to a member of the group, the experience for that kill is zeroed for everyone in the party, not only for the player it is grey to. The effect is that a higher character brought along to speed up the pulls removes the kill experience line for the whole run, so the quest turn-ins and the loot are all that is left. Groups that want both the bar and the bag now form inside a narrow level band, and the spare friend who used to tag along for the clear is a cost rather than a convenience.

In practice this shows up in assembly rather than in the dungeon itself. Finding 5 players within a few levels of each other takes longer than finding 4 plus someone who can carry, so the wait before the first pull stretches and the hours at which a group can be put together narrow to whenever that level band is online. The planning answer is to pick the door by the band you are already in instead of by what the carry could clear, and to decide before inviting anyone whether the run is being judged on experience or purely on what comes out of it, since the same party cannot now do both. The repair and reagent line is deducted either way, which keeps the evening measured against the standing costs of levelling.

The new dungeon sets are the part of the loot line that most groups misread. Forever adds armour sets built around the instances, the Krol'dok Battlegear among them, and the pieces inside one set do not all bind the same way. The shoulders drop bind-on-equip, so they can go straight to the auction house; the helm is bind-on-pickup and is worth only what it does for the character who rolled on it. That split decides which drop is an argument about loot rights and which is simply someone's upgrade, and it is worth knowing before the boss dies rather than after.

Demand for the bind-on-equip pieces is not driven by their stats. Buyers want the look, because a set shoulder is one of the few items a levelling character can wear for appearance alone, and that interest does not fade when the set stops being an upgrade. A stat item sells to players in one level band and then stops; a cosmetic one keeps a buyer pool for as long as the appearance stays wanted. For the ledger it means a single set shoulder can be worth more than everything else the clear produced, which is the strongest argument for treating a dungeon as income first and counting it against the wider Forever gold economy rather than against the experience bar.

So the honest comparison is not only dungeon income against farm income, but dungeon income against the chance of getting no run at all. The practical handling is to treat the attempt as optional rather than as the plan: start a gathering or farming loop, keep the group search running in the background, and switch over only when 5 people are actually standing at the door. A group pulled together from a guild or a standing list carries far less of this risk than one assembled from strangers, which is why the same route can be worth planning for one player and not for another. Either way, the assembly time belongs in the ledger beside the repairs when a run is weighed against the wider Forever gold economy.

The caveat to all of that is that the quest lump does not arrive on the schedule the plan assumes. Blackfathom Deeps and the Stockades both carry sets that do not close inside a single visit: part of the chain turns in outside, the next step points back at the door, and the full reward lands only after a second trip. A group that forms once, clears once and disbands has collected the loot line and roughly half the quest line, which is a different result from the one the ledger was built around.

At the top of the level range the arithmetic changes again, because Dire Maul carries a weekly quest rather than a per-run one. The instance has 3 wings, and the quest pays out up to 100 g on top of whatever the bosses and the trash produce. That is a fixed amount arriving on a fixed schedule, which is something no levelling dungeon offers, and it is the first point where a clear can be planned as income without the loot table having to cooperate.

Below that sits the other addition to the route, Krol'dok Stronghold, built for the 40 to 45 band with 6 bosses in it. The thing worth noting is the gap it fills: that stretch of levels used to be served by instances scattered across several continents, and a single door covering the whole band makes the approach cheaper to plan and the group easier to keep together for a second clear. We do not have drop tables for it and are not guessing at them, so the only honest way to treat the stronghold at the moment is as 6 boss chances at recipes and blueprints plus whatever the trash gives the skinners and the cloth collectors.

Put the 2 together and the back half of the level range reads differently from the front. A character moving through the 40s has a dungeon with 6 bosses to work and, once the level range allows it, a weekly turn-in at Dire Maul that pays whether or not the loot cooperates. That is a route rather than a list of doors: the stronghold for materials and crafting supply on whatever schedule the group can manage, the weekly as the one fixed payment in the plan. The usual deduction still applies first, since repairs, reagents and the training bill for new ranks come out of the same purse before any of it counts as income.

Further down the range, the Alcaz Island prison (48-53) hands the group a decision instead of a drop. Inside the run there is a crate of documents with a vendor price of 25 g, which makes it the one object in the instance with a fixed figure attached before anything has to be appraised or listed. Unlike a bind-on-equip piece it does not depend on who is buying that evening, and unlike cloth it does not need a second player to turn it into money. Sold at the counter on the way out, it is a clean addition to the run total and it covers a fair share of the usual levelling gold sinks on its own.

The complication is that the crate is also the opening step of a chain that the quest log does not advertise, so handing it to a vendor closes that route off for the character who looted it. We do not know what the chain pays out at the end, and we are not going to guess at a figure, so the honest framing is a choice between 25 g now and an unknown reward later. Two things follow for a group. First, decide before the pull who ends up holding the crate, because arguing about it afterwards is worse than splitting the coin. Second, if nobody in the party intends to follow the chain, the crate is simply vendor trash with a good price and should be treated that way rather than carried around for weeks.

Alcaz Prison belongs on the same list and is the newest entry on it. We have no income figures for it yet, so nothing is quoted here: no coin per clear, no resale estimate on what drops, and no view on how its boss recipes compare with the ones further down the level range. What is known is the shape of the thing rather than its payout, and that is enough to plan the first visit around.

The structure that matters for the ledger is the chain attached to it. Entry runs on Fractured Insignia, and the full sequence calls for 6 of them, which means the instance is not a single evening. A group that forms once and clears once has started the chain rather than finished it, and the cost side of the run gets paid 6 times over before the end of it is reached: repairs, reagents, travel and the assembly of the same 5 people on separate nights. Until someone has numbers for what the repeat trips return, treat Alcaz Prison as an unpriced commitment rather than a known earner, and judge it the way you would judge any other long chain against the wider Forever gold economy. We will put figures here when there are figures worth putting here.

Taken together, the 8 November hotfix settles the question this guide has to answer. Dungeon kill experience moved from 40% to 60% of the Classic Era value, Blizzard described the resulting levelling pace as working as intended, and nothing published since points at another pass. A figure that is described as finished can be planned around, so the comparison between an instance hour and a surface hour stops being a bet on the next patch notes and becomes a straight count of what the evening produced.

The count itself does not change with the hotfix. A run is scored on bind-on-equip drops that can be sold to a stranger, on boss recipes and blueprints, and on the cloth and skins that come off the trash, with the repair and reagent bill deducted before anything is called profit. Experience is a side effect of that work rather than the reason for it, and at 60% of Classic Era it is not large enough to cover an empty bag. If a clear returns nothing sellable, treat the hour as a loss and pick a different door next time, because the standing costs of levelling are charged whether the bar moved or not.

Two things follow for the ledger. The first is that the weekly is worth treating as its own task rather than as a reason to run Dire Maul repeatedly: once it is claimed, the remaining visits are back to being judged on drops and materials like any other door. The second is that the wings are not interchangeable in cost, since the approach, the group you need and the time a full clear takes differ between them, so the gold figure should be read against the hour it actually took and against the gold sinks waiting at cap rather than banked in advance. We do not have a breakdown of which wing pays best per hour and are not guessing at one.

The timings players have posted in complaint threads are worth writing into the plan, because they are longer than most groups assume when they sign up. A Gnomeregan clear was reported at 1 hour 40 minutes, a Drowned City run at 1 hour 53 minutes, and one player needed 4 separate trips through the Scarlet Monastery graveyard wing before a single quest objective closed. Another chain cost 2.5 hours of prerequisite questing on the surface before the instance step could be started at all. These are single reports rather than averages, and group composition and route knowledge will move them in both directions, but nothing in them suggests a short evening.

Converted into the only unit that matters, a clear in that range plus assembly, travel and the walk back is a 2 hour block of play. Everything that comes out of it, the bind-on-equip drops, the boss recipes, the cloth and the skins, has to beat 2 hours of a gathering route on the same character, with the repair and reagent bill deducted first. The 4 runs for 1 objective are the sharper case: when a quest reward is spread across 4 clears, the lump at the turn-in divides by 4 before it joins the hourly figure, and a prerequisite chain that eats 2.5 hours before the door opens is part of the same sum. Count the whole block, not the pull timer, and weigh it against the recurring costs of levelling that run whether the group forms or not.

That leaves the question of why anyone goes back to the same door a 3rd or 4th time, once the quests attached to it are spent. The answer is no longer the bar. A repeat clear pays in bind-on-equip drops, in the recipes and blueprints that bosses hold, in the cloth and skins the trash leaves behind, and in the pre-raid gear pieces that a character would otherwise have to buy at market price. None of that is experience, and none of it should be written into the plan as experience. A farmed dungeon is a supply run with a boss list attached.

Below level 30 the quest line is where most of that top-up sits, and the 3 doors worth checking before a group forms are these:

  • Ragefire Chasm, flagged for characters around 9-19. Short, central, and the usual first instance for a Horde character who has not yet left the starting cities behind.
  • Hall of Thanes, flagged for characters around 11-19. The Forever addition in that band, and the one most groups have least practice with.
  • Ruins of Lordaeron, flagged for characters around 15-24. The widest band of the 3, which means a mixed group can usually all take something from it.

2 details decide whether the lump arrives. Part of this set is faction-specific, so a quest a guildmate describes may not exist for the character standing at the door, and the Horde and Alliance lists for the same level band are not interchangeable. Part of it is picked up inside the instance rather than from a giver outside, which means a group that sprints to the first boss can walk past the item or the corpse that starts the chain and never learn it was there. Collecting the outside givers first and pulling slowly through the opening rooms is the whole technique, and it is worth the extra minutes against the running costs of a levelling character.

The next step up from the low brackets is The Drowned City, which covers roughly levels 37 to 42. The entrance sits on the coast of Stranglethorn Vale, so the approach is a travel cost in its own right for anyone coming from a capital, and that time belongs in the ledger before the first pull. It carries 5 quests. As with the earlier doors, the set is not handed out at the instance portal: part of it is picked up away from the coast and only points back at the door afterwards, which means a player who rides straight to the entrance and queues in will be running the loot line alone. We do not have a confirmed giver-by-giver list and are not going to guess at one, so treat the pre-run step as a search rather than a known route and allow for it when the group agrees a start time.

What we cannot give is money. There are no reward figures for the 5 quests that we are willing to put in print, and nothing published that would let anyone work out what a full clear with the set in the log returns in coin. Until that exists, the only honest way to judge a Drowned City evening is the same way the lower brackets are judged: count the bind-on-equip drops, the boss recipes and blueprints, and the cloth and skins off the trash, then deduct the repair and reagent bill and the ride down the coast. If the bag is thin, the quest turn-ins may or may not cover the difference, and that uncertainty is worth stating to the group before they commit the hour rather than after. The fixed part of the arithmetic is still the standing costs of levelling, which do not pause while a 5-person group assembles.

Judged that way the decision becomes simple arithmetic. Count the bag at the end, deduct the repair and the reagents, and set the result against what the same hour would have produced on a gathering route or at a crafting station. If the group is going back for a specific recipe or a specific drop, the run has a target and can be abandoned once the target lands or clearly will not. If it is going back out of habit, it is competing with every other way of earning in the Forever gold market, and it usually loses. Repeat visits are worth planning around the items you actually need, not around the number of clears.

The same pattern appears in the quests that ask for an item rather than a kill. Several of the chains attached to Blackfathom Deeps, Zul'Farrak and the Drowned City want something that comes off the trash instead of off a boss, and a trash drop is not guaranteed by clearing the room it sits in. A group can pull everything on the way through, finish the instance clean and still walk out without the object in the bag, which leaves the turn-in open and the lump unpaid until somebody goes back in. In part of that set the quest only opens once the item has appeared, so the first visit is the one that starts the chain and a later visit is the one that closes it. We have no drop rates for any of them and are not inventing figures, but the structure is enough to plan around.

For the ledger this means the second trip is not a half-priced version of the first. The travel, the assembly and the repair bill are paid again in full, while the loot line on a repeat clear is whatever the same bosses happen to give a second time. The practical move is to treat an item quest as a reason to hold the group together for another pull on the same evening, because re-forming 5 people later for one drop is the point at which most chains are quietly dropped. If the party does split, count the return trip as its own hour and judge it against a surface route, with repairs and reagents deducted twice rather than once, since the gold sinks behind levelling do not scale down for a short visit.

The behaviour data behind all of this points the same way. Around 83% of players enter a given dungeon exactly once and never return to it, which means the typical instance is a single visit rather than a route that gets run down. That reframes the whole calculation: a door is a one-off payment, not a repeatable income stream, and anything that only pays on a second trip is in practice unpaid for most of the people who walk through it. Planning a levelling budget around repeated clears of the same instance describes what a small minority actually do.

It also explains where the money in the 20 to 30 band comes from. Across that stretch the gold a character accumulates is mostly quest reward money, with dungeon loot and vendored trash sitting underneath it rather than carrying it. If the quests attached to an instance are not in the log, the single visit most players make returns the materials line and little else, and the lump that would have made the evening worth assembling 5 people for is simply not collected. Budget the band as quest income first, treat each dungeon as one payment rather than a farm, and deduct the fixed costs of levelling from what is left.

The second trip is the harder half, because the limit on dungeon income here is not the clear but the assembly. Few groups sit in the finder at any given hour, and a party put together for one run rarely survives long enough to repeat it, so the realistic unit of planning is one door per evening rather than one chain per evening. Measured honestly, a dungeon hour includes the waiting, the summons and the walk back, and a surface route that starts the moment you log in will often beat it on gold per hour simply because nothing has to be arranged first. That is the trade to weigh: the instance pays better per hour spent inside it, and worse per hour spent getting there.

The common reading of why the experience was cut at all is that it was aimed at boosting rather than at ordinary groups. A geared character pulling an entire instance while a low level stands at the entrance was the fastest levelling service anyone sold, and it was the service most often paid for with real money rather than with gold. Moving the bulk of the reward out of kills and into quests breaks that arrangement at the root: the passenger has to collect the chain, be present for the turn-ins and make the second trip, so the thing being sold is no longer a level. We have no statement of intent to quote and are not treating the motive as confirmed, but the effect is visible in the channels, where offers of paid levelling runs have largely stopped appearing.

A hotfix has since moved the kill side back up. Experience from mobs inside instances was raised, which narrows the gap between a clear done with the quest chain in the log and one done without it. Nothing in the money line changed with it: the coin on the corpses, the cloth and skins off the trash, and the bind-on-equip drops are all unchanged, so a run is worth exactly what it was worth before in gold terms. Only the levelling half of the comparison moved, and we have no published figures for how far, so the sensible reading is that kills matter more than they did immediately after the cut and still less than they did at launch.

Forum opinion has not followed the hotfix. The common complaint is that the adjustment does not fix the part that actually costs time, which is the assembly rather than the pulls, and that an hour spent finding 4 other players, walking in and clearing still returns less than an hour spent questing on the surface. That argument holds whichever way the experience moves, because it is a judgement about the wait rather than about the reward, and it is the reason groups form for a specific door and a specific drop rather than for levelling. Judge the run on what comes out of it in materials and sellable items, deduct the recurring costs of levelling first, and treat the experience as whatever lands on top.

For a player who never bought one, the knock-on is still worth understanding. Group levelling thinned out because the sold version of it vanished along with the free-riding version, and what remains forms for loot, recipes and materials instead of for the bar. It also removes a familiar reason to hold a large gold balance early, since nobody is saving up to pay a carrier, and it pushes the demand for gold back towards training, repairs and crafting supply. Judge an instance on what comes out of it in items, and the shape of the evening makes sense; judge it on levels gained, and it will keep looking worse than it is.

There is a further gap between the figure a plan produces and the one a run returns, and it comes from how groups behave rather than from what drops. A party assembled for a dungeon is usually 5 people with 5 different objectives in the log, and once the last of those objectives ticks over, the reason to stay ends. Someone has the kill they came for, someone else has the item, and the group thins out before the wings nobody needed are touched. Full clears are the exception, not the normal shape of an evening, and a partial clear takes the bind-on-equip chances, the cloth and the boss recipes down with it.

On 8 November a hotfix raised experience from mobs inside dungeons by 50%. That moves the one quantity the comparison above was built on: the kill line, which had been the weak half of a clear, now carries a larger share of what an instance hour returns, and the quest turn-ins stop being almost the whole case for going. The loot line is unchanged, so nothing about bind-on-equip rights, cloth or boss recipes shifts with it. What shifts is the answer to the question of whether a door is worth assembling people for when the bags matter less than the bar.

In practice the effect lands hardest on the 13-40 band, where the dungeon list is dense and a surface gathering route is still the obvious alternative for the same hour. Before the hotfix the honest advice was to judge a run as a gathering hour unless the quests were in the log; after it, a clear with no quests attached is no longer a dead evening, because the trash itself pays something towards the level. The arithmetic that does not change is the overhead: waiting for a group, the summon and the walk back still come out of the hour before anything is counted, and the repair and reagent bill is still deducted first, which is where the standing costs of levelling decide whether the run was worth arranging. A route you can start the moment you log in keeps that advantage. The hotfix narrows the gap rather than closing it, and we have no figures for where the line now falls, so the comparison is still worth making per character rather than assumed.

Players discussing the change describe a second effect on top of that, and it is about who signs up at all. Once a door has been cleared, few people want to walk it again, because the repeat pays the loot line only and the quest lump is already spent. Tanks and healers are the first to stop queueing, since they are the roles that carry the setup cost of an assembly they gain least from, and a group that cannot fill those 2 slots does not form. The reported workaround at low levels is to run stacks of a single class together, which keeps the trip moving without arranging roles and leaves nothing to argue over when a bind-on-equip drops, because nobody else in the party can use it.

One figure from the same discussion gives the scale people are working from: 3 runs of Razorfen and ROL on a level 17 character returned roughly 8k experience per run, with each run taking about an hour. Whether that reads as good or poor depends on what the alternative hour produces, but the behaviour it drives is the part that matters for the auction house. Fewer repeat clears means fewer bind-on-equip pieces, fewer boss recipes and less dungeon cloth reaching the market, so the supply of instance loot is thinner than the number of levelling characters would suggest. If you craft from those recipes or resell that gear, plan for intermittent stock rather than a steady feed, and deduct repair and reagent bills before calling a run profitable.

The other half of that behaviour is where the levelling crowd went instead. With kill experience inside instances cut, repeating the same door is no longer the obvious way to move a bar, and the players who used to chain runs now form pull groups in the open world and level on stacked mobs there. That is a different activity with different requirements: it needs area damage and a healer who can hold a pile together, not a tank, a key and a full set of dungeon quests. The effect on the finder is immediate, because the population that used to keep it stocked is doing something that does not involve queueing at all.

For dungeon income this cuts both ways. Assembly gets slower and less reliable, so the waiting counted into a dungeon hour grows and the comparison with a surface route tilts further towards the route. At the same time, fewer clears means fewer boss recipes and blueprints reaching the auction house, and fewer bind-on-equip pieces off the trash, so what does come out of an instance meets thinner competition when it is listed. A player who is willing to arrange the group anyway is supplying a line that most of the server has stopped feeding, which is worth more than it was when everyone was in there for the experience. The bill is unchanged either way: repairs, reagents and the recurring costs of levelling come off the top before any of that counts as profit.

That also changes what makes a player queue for the same door a second time. With the experience motive gone, nobody repeats an instance to finish a level band; they repeat it because something specific inside it is worth owning or selling. In practice that means a named bind-on-equip piece, a recipe or blueprint held by one boss, or a steady cloth and skin return that happens to beat the surface route for that level range. If none of those apply, a second visit has no case at all, and the group will not form however easy the clear was the first time.

The planning version of that is to treat each door as a short list of wanted items rather than as a block of time. Write down what the instance can hand over that is actually sellable or usable, decide which boss carries it, and accept that the wings holding nothing on the list are optional. A run built that way ends when the list is cleared instead of when the last room is, which matches how groups already behave and keeps the repair and reagent bill proportionate to what the bags come out with.

The same behaviour shows up across weeks rather than within one evening. When experience was the main reward, a door was worth repeating until the level band was finished, and the same instance could carry a character for several sessions. With the loot line doing the work instead, a player goes in for a specific thing: a recipe or blueprint off a named boss, a bind-on-equip for a slot that has been empty since the last zone, a stack of cloth for a tailoring push. Once that thing is in the bag, the reason to queue again is gone, and repeat traffic through the same instance falls off quickly.

That matters for the comparison the rest of this guide is built on. A gathering route can be run again tomorrow on the same terms, because the nodes return and nothing has to be arranged. A dungeon pays best on the first visit and less on each one after it, while the cost of assembling 5 people stays flat or gets worse as fewer players want that particular door. The practical reading is to treat a run as a one-off errand with a named target rather than a repeatable income source, to go in with the quests already in the log so the trip is not wasted, and to keep the surface route as the thing that fills the hours between those errands.

So the per-level numbers on this page should be read as a ceiling for a complete run, not as an expectation. The practical correction is to treat anything beyond the quest objectives as optional income that may or may not happen, and to decide before the first pull whether the group intends to finish the instance or only the log. If the answer is the log, the honest comparison is against a gathering hour again, because the loot line will be short and the repair and reagent costs of levelling are deducted from the shorter total all the same.

If the run is no longer levelling, the honest way to judge it is a short ledger taken at the door on the way out. There are 4 lines worth writing down: what the bind-on-equip drops will list for, what the stacked materials are worth once the cloth and skins are sorted, what the recipes and blueprints add if any dropped, and what the rest vendors for after repairs and reagents are deducted. Nothing in that list is time to the next level, which is the point. A clear that produced 2 sellable items and a bag of cloth is a result; a clear that produced a full bag of vendor trash and a repair bill is not, however smoothly it went.

Gathering nodes inside an instance belong on the same ledger when they are there, because they are uncontested in a way that an open-world node never is and nobody rides past you to take one. We do not have a list of which of the 9 dungeons hold ore or herb spawns and are not guessing at one, so the practical approach is to bring a gatherer on the first clear and note what the route actually passes. Counted that way, a dungeon evening sits next to a surface route as 2 comparable ways of filling the same bags, and the comparison decides whether the group forms again tomorrow or whether the hour goes somewhere else. The same ledger is what tells you whether the night covered the standing costs of levelling or quietly fell short of them.

What is not settled is who gets a node when the group walks past one. There is no loot rule covering ore and herbs the way there is for items off a boss, so the question is handled socially and handled differently by every party. Some groups treat a vein as belonging to whoever has the skill and reaches it first, some expect the gatherer to ask before stopping, and some never discuss it until somebody mines without asking. Players report both outcomes of that silence: parties where a miner was removed from the group for taking a vein the leader considered shared, and parties that settled it with a /roll in chat the same way they would for a bind-on-equip drop.

For the ledger above that matters, because a removal mid-clear costs the whole evening and not just the node. The cheap fix is 1 line at the door before the first pull, asking whether nodes are free for the gatherer or rolled on, and then holding to whatever the answer was. If the group says rolled, the stack comes out split and smaller than a solo route would give; if the group says free, the uncontested spawns are the main reason to bring the gatherer at all. Either answer is workable. Finding out which one applies after you have already mined is the version that ends the run early.

The coin line deserves a harder look than it usually gets. Players report boss loot in levelling instances vendoring for 1 copper a piece, which means the items that cannot be worn and cannot be sold to another player are worth nothing at all on the way out. A clear that drops nothing wearable and nothing bind-on-equip therefore ends with a repair bill and a reagent bill against a pile of 1 copper vendor trash, and if the group wiped on the way the run finishes in the red. That is not a rare outcome, it is the default one whenever the drop tables do not cooperate, and it is why the materials line above carries more of the run than the loot line does.

The ledger also has to start earlier than most people start it. There is no matchmaking tool here, so a group is put together by hand in chat, and the time that takes belongs to the run. Players describe a Hall of Thanes party costing roughly an hour to assemble before anyone moved towards the door, between finding a tank, waiting on the last damage slot and summoning the people who were somewhere else. Counting the clear at 45 minutes and ignoring the hour in front of it produces a number that has nothing to do with the evening you actually spent.

Counted honestly, gold per hour runs from the moment you start looking, not from the first pull. That changes which doors are worth opening: a dungeon with a thin loot table is still workable if the group is already standing there, and a rich one is a poor use of the night if it takes longer to fill the party than to clear it. It also argues for keeping a group together for a second and third run rather than disbanding at the exit, since the assembly cost is paid once and then spread across everything that follows. If the whole block of time, search included, does not cover the standing costs of levelling, the problem may be the waiting rather than the drops.

The other half of the bill is time. One account of a Deadmines evening puts 45 minutes on finding and assembling the group before anybody walked through the door, which has to be counted against whatever the bags hold at the end. An hour of gathering produces something sellable every time out; an hour that is mostly spent in chat looking for a tank produces nothing until the group exists. If the evening is meant to cover the standing bills of levelling or put coin towards the riding bill at 40, the assembly time belongs in the calculation at the same rate as the run itself. We have no figures for how long a group takes to form on average and are not inventing them.

Reports from testers describe a group finder that assembles the party but does not move it, with no teleport to the instance door once the group is full. If that holds, the assembly time above is only the first half of the delay: every member still walks or rides to the entrance, and the slowest arrival sets the start of the clear. We treat this as unconfirmed, since it comes from player accounts rather than from anything documented, and it is worth checking yourself before planning an evening around it.

There is a third piece of time that the ledger usually misses. Part of the quest set attached to the second Horde dungeon, the Ruins of Lordaeron, is not handed out at the door: it sits behind quest chains that open from level 13, and a character who walks in without having run them will clear the instance and collect only the loot line. The quests are not lost, but they are not free either, because the chain has to be worked through on the surface before the dungeon turn-ins become available at all.

For the comparison above that means a dungeon evening on the Horde side can carry 3 time costs rather than 2: the chain, the assembly, and the clear itself. Counted honestly, the first clear of that door is the expensive one and later runs are cheaper, since the chain is done once per character. We have no figures for how long the chains take and are not inventing any, so the practical approach is to run them on a quest night when you are in the area anyway, and judge the instance against a gathering route only once the quest access is already paid for. Either way the result still has to cover the standing costs of levelling before the night counts as profit.

The same accounts add that meeting stones outside the doors do not summon anybody, so the group has no way to shorten the approach once it is assembled. One player describes roughly 15 minutes of riding to reach the entrance after the party was already full, with everybody else waiting at the door for the last arrival. We cannot confirm that the stones stay inactive at launch and we are not treating the 15 minutes as an average, but both reports point the same way: the walk is part of the run, not a preliminary to it.

Practically that means the approach comes out of the ledger before anything is counted as profit. If assembly took 45 minutes and the ride took another 15, the clear itself has to carry an hour of overhead on top of its own length, and the bags at the end are judged against that total rather than against the time spent inside. Two useful habits follow from it: pick the door nearest to wherever the group already is instead of the one with the better drop table, and bring the character that can get there fastest, which is one more argument for budgeting the riding training bill at 40 earlier rather than later. A run that returns a decent bag still loses to a gathering hour if half the evening went on travel.

One change pushes traffic the other way. A token off a raid boss does not hand over the finished piece: developers have described it as the start of a chain, and the chain sends the holder back into group content rather than to a vendor. That means the dungeons above are not only a levelling-band activity any more. Characters at the level cap who already cleared them years ago will be forming groups for the same doors, because the tier piece is gated behind steps that cannot be finished alone.

For anyone judging a dungeon hour, this is a demand shift rather than a loot change. The same instances now compete for tanks and healers across 2 populations at once, which lengthens assembly time for levelling groups and puts a price on being the one who is already geared. It also creates obvious ground for paid runs: a raider who needs a specific step finished will pay a group to carry it rather than wait, and that payment comes out of the same wallet that covers the repair and travel costs of levelling. We have no figures for how many steps the chain contains or which dungeons it touches, and are not guessing at them.

One caveat applies to every time-based number on this guide. The final patch notes of 2 November raised class damage and increased the health of higher-level mobs, and both changes move the clock on a clear in opposite directions. Figures that testers wrote down on beta before 15 October were measured against the older tuning, so any gold or materials per hour derived from them is a starting estimate rather than a result. The ledger method still holds; the minutes attached to it do not, until somebody times the same doors again on live.

In practice that means treating the lower-level instances and the higher-level ones differently for now. Where the damage increase dominates, a clear should finish faster than the beta notes suggest and the hourly figure improves on its own. Where the extra mob health dominates, the same run takes longer and the repair and reagent lines grow with it, which is enough on its own to push a marginal door below the standing costs of levelling. Re-time one familiar clear after the start, write down the bag at the end, and use that instead of anything carried over from beta.

The practical effect is that travel is not a one-off cost paid at the start of the night. A group that breaks up after 2 clears and reforms with a different member pays the approach again, and the new arrival may be starting from a flight point on the other side of the zone. Where a teleport would collapse that to nothing, the current arrangement puts it back into the hourly figure, which is the main reason a dungeon evening can return less per hour than the loot on the floor suggests. Compare it against a gathering route on the same clock, with the approach counted in, rather than against the contents of the bags at the end.

It helps to know how many doors are actually in play. The announced list runs to 9 new dungeons covering levels 13 to 60, split across the 2 continents:

  • Eastern Kingdoms: Hall of Thanes, Ruins of Lordaeron, Excavation Site, City of Dalaran, The Drowned City, Krol'dok Stronghold.
  • Kalimdor: Alcaz Prison, Blackmaw Hold, Shaper's Terrace.

The 28 to 34 band is where the City of Dalaran sits, and it is the largest of the new doors in that range. It carries 8 bosses, which matters for the loot line more than the level range does: 8 kills is 8 rolls on recipes, blueprints and bind-on-equip drops rather than the 3 or 4 a short instance offers, and the materials off the trash between them stack up accordingly. The trade-off is clear length. A run with 8 bosses takes longer than the band's smaller instances, so the per-hour figure only holds if the group is stable enough to finish rather than breaking up after the first half.

There are 4 quests attached on the faction side, which pay out once per character and should be taken on the first visit rather than left for a later clear that may not happen. There is also a secret boss, and reaching it requires a preliminary chain that runs through Shadowfang Keep, so the encounter is not something a group stumbles into on a normal evening. Treat that chain as separate work: the trips it takes produce whatever Shadowfang Keep itself drops, and the secret boss is a bonus at the end rather than the reason to go. We have no information on what it drops and are not guessing, so plan the night on the 8 regular bosses and count the detour against the standing costs of levelling like any other hour.

That spread matters for income because the same run means different things at the bottom and the top of it. A clear in the low teens pays in cloth of the cheapest tier and vendor trash, while an instance near 60 sits in the band where bind-on-equip drops have a real market and recipes are worth listing rather than vendoring. The range also straddles the riding bill at 40, which is the point where most players start judging an evening by what it contributed to a specific purchase rather than by whether the bags looked full. Hall of Thanes is the one we can describe in detail; for the other 8 we have the names and the level band and nothing on drop tables, trash density or clear length, and we are not filling those gaps with guesses.

One more thing belongs next to the Hall of Thanes income figure: by the statistics collected during the beta it was the deadliest of the instances on test, measured by player deaths rather than by clear time. That is a 13 to 18 door, so the group walking into it is doing so with low-tier gear and a thin repair budget, and the deaths land on the same characters who are counting coppers for the next bag or the next set of abilities.

The practical effect is that the run total quoted for a clear is a gross figure, not a net one. A wipe costs durability on every piece worn by every member, plus the walk back from the graveyard, plus whatever reagents were spent on the pull that failed. We have no durability or repair figures for this instance and are not inventing any, but the direction is clear enough: the more often a group wipes, the more of the clear goes back over the vendor counter, and a night that ends 2 or 3 wipes deep can hand most of the coin straight to repair bills and other gold sinks. If you are judging whether the instance is worth the evening, subtract the deaths from the loot before comparing it against anything else.

The split itself is uneven and that has a planning consequence. 6 of the 9 doors sit in Eastern Kingdoms and 3 in Kalimdor, so a character levelling on the eastern continent has more instances within reach of the same flight network, while a Kalimdor character is more likely to cross water to reach the next one in the band. Nothing about that changes what a clear pays; it changes how much of an evening goes into the approach before the first pull, which is the part that quietly decides whether a dungeon night beats the gold sinks of levelling or just keeps pace with them.

The names carry hints and not much more. Ruins of Lordaeron, City of Dalaran, The Drowned City and Excavation Site point at places anybody who has levelled through the eastern zones can picture, and Alcaz Prison names an island most players have swum past at least once, but we have no confirmation of where the entrances actually stand, which level each instance opens at, or how the band from 13 to 60 is divided between them. Hall of Thanes is documented as the 13 to 18 door and is the only one we treat as settled. Until the rest are run and written up, the sensible use of the list is as a count of how many instances exist in the levelling range, not as a route plan.

On a PvP ruleset the approach has a second cost that does not show up in the repair bill. A forum summary put together by players sorts instance entrances by which faction can reach them through friendly territory and which has to cross contested ground to get there, and the two lists are not the same length for Alliance and Horde. The point of that summary is that a door standing in your own faction's zone is reached at walking pace, while a door on the far side of a contested zone means the slowest member of the group may arrive twice, or not at all, and the clear starts late either way. None of this is a drop-rate question; it is simply which entrances you can plan an evening around without the ride being part of the gamble.

Two things are worth saying plainly about that list. It is a player compilation drawn from accounts of actual runs, not an official breakdown and not anything taken from patch notes, so the groupings reflect what testers found on their own characters and may not hold for every zone or every level range. And it says nothing about what the dungeon pays once you are inside. Use it the way you would use a route note: if 2 instances return a similar amount for similar time, the one your faction reaches without crossing contested ground is the cheaper evening, because the minutes lost to a bad approach come out of the same column as the loot.

All of the above assumes the run is paying twice: once in loot and once in levels. When the experience from mobs inside an instance is cut, the second payment shrinks and the clear has to stand on the loot line alone. That is a harder test than it sounds. A group that was willing to spend time on assembly and approach because the pulls were also moving the level bar now has to justify the same time against cloth, skins, a few recipes and whatever bind-on-equip drop the tables hand out, and the coin on the corpses still vendors for what it vendored for before. Nothing about the income changed; what changed is that there is less on the other side of the ledger to absorb a bad night.

The second and third clear of the same instance make that gap plain. There is no repeat bonus attached to running a dungeon again on the same character: the drop tables roll as they did the first time, the coin on the corpses is the same coin, and nothing pays out for having cleared the place before. So a group that stays together for a third run is doing it for the chance at a bind-on-equip drop and for the cloth, not because the run itself is worth more than the one before it. Measured against the assembly and approach time above, that is a thin reason to go back in, and it is why most evenings end after the group has what it came for rather than after the clear that would have paid best.

Players have put forward their own fixes for this in discussion threads, and the shape of them is consistent. The common suggestion is a small amount of Merchant's Favor on each boss, in the range of 1 to 2 per kill, so that a repeat clear returns something countable even when the tables give nothing wearable. A second version drops the currency entirely and hands out tokens that are traded for loot from the same instance, which keeps the reward tied to the dungeon rather than to the market. The argument against paying repeats in gold or in raw materials is the one worth noting: both go straight into the gold sinks of levelling and the auction house at the same time, and a reward that scales with how many times a group can re-enter pushes supply and prices around in ways a fixed drop table does not. None of this is announced, and we are not treating it as planned; it is a player proposal and nothing more until something official says otherwise.

The visible result is players moving back out into the open world, where quest turn-ins, gathering and ordinary overland kills keep both lines intact. That shift is worth noting when reading any older estimate of what a clear is worth, because those estimates were written when the levels came free with the loot. Treat a dungeon under those conditions as an income activity and compare it directly with an hour of gathering: if the bags at the end do not beat the hour, the run was a social evening rather than a source of gold. We have no figures for the size of the experience change or for how long the current arrangement stands, and are not inventing them.

The complaint is coming from beta testers rather than from anyone reading documentation, and the shape of it is consistent: instance clears are returning less experience than the time spent inside them led players to expect. The objection is not about loot tables or drop rates. It is that the run takes the same minutes it always took while paying less on the second line, so the argument that used to settle the question, that a dungeon is slower in coin but faster in levels, no longer settles anything on its own.

The same accounts describe where that experience went: into the quests attached to the instance rather than into the mobs inside it. A clear done for the first time still pays well, because the turn-ins land at the end of it, but those turn-ins are one-time. The second clear of the same dungeon pays the quest portion exactly once and nothing after that, so a group that used to chain 3 or 4 runs in an evening is paid at full rate for the first and at loot rate for the rest. We have not seen this confirmed outside player reports and are not treating it as documented.

For anyone planning an evening around dungeons, the practical consequence is that a repeat run has to be judged purely as an income activity: cloth, skins, recipes, whatever bind-on-equip drop appears, and the coin on the corpses against the repair and reagent bill. That also changes who is in the group. A player who already has the turn-ins behind them has less reason to sit through assembly and approach for a dungeon they have cleared, which thins the pool of tanks and healers available for the people still on their first pass. If the group is slow to form, that is usually the reason rather than the time of day.

There is a ceiling at the top of this that is worth stating before anyone plans a levelling route around dungeon income. At 60 there is no heroic version of a 5-man to step up into, so the same instances pay the same way they paid on the way up: cloth, skins, the occasional recipe, a bind-on-equip drop when the tables cooperate, and vendor coin that barely covers the walk back. The run does not get harder and it does not get richer. Whatever a clear was worth while levelling is roughly what it is worth afterwards, except that the experience line is gone entirely and the loot line has to carry the whole evening on its own.

Players have been asking for the obvious fix, which is a harder tier of 5-man that returns more gold and more materials for the extra difficulty, on the grounds that a group willing to take a sharper fight should see it in the bags at the end. We have no indication that such a tier exists or is planned, and we are not going to describe one that has not been shown. Plan around what is there: at 60 the dungeon is a supply line for crafting materials and recipes rather than a way to cover the standing gold sinks at 60, and if the evening has to pay for repairs and consumables as well, compare it honestly against an hour of gathering before committing the night to it.

The stated plan among several of those testers is to skip instances entirely through the early levels and come back to them somewhere in the 40s, once the gear on offer is worth the detour on its own. That is a personal decision about levelling speed, but it has a direct effect on everybody else's income, because a dungeon is the one activity on this list that cannot be done alone. Gathering does not care how many players agree with you. A 5-man clear only happens if 4 other people in the right level range reached the same conclusion about the same instance on the same evening.

If that intention holds in any volume, the assembly time described above gets worse rather than better at the low end, and the effect is heaviest on whichever role the group is short of. The run itself does not slow down; the wait before it does, and the wait is paid at the same hourly rate as the clear. Two things follow for anyone counting a dungeon as income. Check what the level bracket actually looks like before committing an evening to it rather than assuming the group will appear, and treat the early instances as something you do when a group already exists rather than something you plan a night around. The brackets most likely to hold a party are the ones where the gear is worth travelling for, which is also where players saving towards mount training at 40 tend to concentrate. We have no numbers for how many players intend to skip, or for how much the wait grows, and are not inventing them.

There is a second side to this that the loot arithmetic above does not touch. A dungeon is not only something you run for your own bags; it is also something players sell to each other. A geared character clearing an instance for a lower-level buyer is a paid service, and the price of that service is set by what the buyer gets out of the run, which is mostly levels. When the experience inside instances drops, the seller's hour is worth less to the buyer, and the rate the service can be asked for moves with it. That is why the same change shows up in 2 different arguments on the forum: one from players levelling their own characters, and one from players whose gold comes from running other people through.

It also explains why requests to revisit the experience numbers keep reappearing rather than settling. The people asking are not all asking for the same thing. A levelling player wants the clear to be worth the assembly time again; a seller wants the service to be worth offering at all. Both are arguing about the same line in the same ledger from opposite ends of the transaction. What none of those threads contain is a figure: the discussions describe the effect in terms of runs feeling slower and services being harder to price, not in terms of a stated percentage or a documented adjustment. We have not seen concrete numbers attached to the change, and treat any that circulate as unverified until something official says otherwise.

Players putting a number on it estimate instance experience at roughly 1/5 of what the same time spent on quests returns. That figure comes from testers comparing their own evenings rather than from anything published, so treat it as a rough reading and not a measurement, but it matches the behaviour described above: groups that used to accept a slow clear because the level bar was moving no longer accept it. At 1/5, the levelling argument for a dungeon stops working on its own, and nothing is left on that side of the ledger to cover a bad night of drops.

The practical consequence is a change in what a run is measured by. A clear judged on levelling speed was always compared against questing in the open world, and the comparison was about minutes per level. A clear judged on loot and gold is compared against a gathering route, and the comparison is about what the bags hold when the group breaks up. Those are different tests with different answers, and the second one is harsher, because the assembly time, the ride to the door and the repair bill all come out of the same column as the income. If the evening is meant to put coin towards riding training at 40, run the loot test and ignore the level bar entirely.

The sharper version of the same complaint concerns what happens after the quests inside an instance are handed in. Testers describe mobs in a cleared dungeon paying no experience at all once the quest line attached to them is finished, which would mean a repeat clear moves the level bar only through the quest rewards it no longer has. We cannot confirm that from anything documented and are not treating it as settled, but the behaviour that follows from it is already visible: groups stop clearing and start skipping, pulling only what blocks the corridor and heading for the boss that holds the drop they came for.

That changes what this page is measuring. A run assembled to reach 1 boss is not being judged on levels, because there are none on offer past the first visit; it is being judged on what the boss drops and how many minutes the approach costs. The 2-line comparison above collapses back to 1 line, and it is the loot line. Under those conditions the same tests apply with less margin for error: count the assembly and the ride to the door, count the repair and the reagents against the recurring costs of levelling, and compare the bags at the end with an hour spent gathering. A boss skip that pays is worth repeating. A boss skip that does not is a trip taken for the item, which is a reason to go, but not an income one.

There is a ceiling on all of this that the arithmetic above quietly ignores. Players are discussing a cap on how many instances a character or account may enter within an hour, which puts a hard stop on the repeat-clear plan long before the group gets bored. If such a cap exists, the hourly figure for dungeon income is not set by how fast the group clears but by how many entries the hour allows, and a fast group simply finishes its allowance early and waits. That matters most for the boss-skip pattern described above, since skipping is designed to raise the number of runs per evening, which is exactly the number a cap limits.

Both halves of this belong in the same warning. The entry cap and the reduced experience inside instances come from forum discussion among testers, not from a published patch note, and we have seen no official figure for the number of entries permitted, the length of the window, or the size of the experience change. We are not inventing any of them. The practical response is to treat a dungeon evening as capped until proven otherwise: plan the route and the gathering around it, and check the limit yourself on the character you intend to farm with rather than on a number repeated in chat.

For the planning this page is built on, that turns a 1-line comparison into a 2-line one. A gathering route was always the alternative on gold; under the current arrangement it is also the alternative on levels, since overland kills and quest turn-ins along the route keep both columns moving. Before committing an evening to clears, run the test both ways: what the bags hold at the end against an hour of gathering, and what the level bar moved against the same hour spent outside. A dungeon that loses on both is a social evening, which is a fine thing to spend the night on as long as it is not being counted as income.

Travel time belongs in the same column, and it is the part most old notes get wrong. Several of the shortcuts players learned years ago no longer work in Forever: gaps in the terrain that used to let a character drop through and come out closer to the door have been closed, the path under the tower at Medivh's has been shut, and the approach used to reach Upper Blackrock Spire without the long way round is gone. Routes that assumed any of those are now longer than the note says, and in some cases they do not exist at all.

The working rule is to time the approach on the current geometry rather than from memory. Walk the route once, count the minutes from the nearest safe point to the instance door, and use that number when deciding whether a clear fits inside the window you have. A run that looked cheap because the entrance was 2 minutes away can turn into a different proposition when the only remaining path is the overland one, and that difference lands on every clear of the evening rather than on the first one. We have no full list of which shortcuts were removed and which still hold, so the only reliable figure is the one you measure yourself.

Faction decides how much of that approach you pay for. Deadmines sits on the points list of starter instances the same way for everybody, but the two sides do not arrive on equal terms. Alliance characters reach it from a town they are already questing out of. Horde characters take a zeppelin, land on the wrong side of the continent and finish the trip with a swim, and the water route passes through a zone built for much higher levels than the instance is. The clear is the same clear; the bill in front of it is not.

The second half of that gap is what sits around the door. For a character levelling nearby, a run folds into work that was happening anyway: turn-ins waiting outside, objectives inside the instance that complete on the same pull as the loot. A Horde character crossing to Deadmines has none of that, so the whole trip has to be paid back by what the bags hold at the end. In practice that means the instance is worth travelling to when the group is already formed and intends to clear several times in a row, and is hard to justify for 1 run. The same test applies anywhere the entrance is on the other side of the map from where your quests are: measure the approach once, then decide how many clears it takes to make the approach worth repeating.

The mirror case runs the other way. The early Kalimdor instances sit inside zones the Horde levels through, so an Alliance group reaches them by crossing ground where the other side is questing in numbers. On a normal ruleset that is only distance. On a PvP ruleset the approach stops being a fixed figure and becomes a range: the same road takes one length when nobody is on it and another when it is busy, and the second case is the one that decides whether a clear fits in the evening.

The cost of that is not only minutes. A death on the way in means a corpse run before the first pull, durability loss on gear that has not earned anything yet, and a repair you cannot make until the group reaches a vendor it can actually use, which may be the town you started from. Parties in that position often clear with damaged gear rather than walk back, and the bill lands later alongside repair bills and other sinks. The practical answer is the same as for travel time: treat the approach as part of the run, count it on the ruleset you play rather than the one the old note was written on, and plan for several clears in a row when the road is the expensive half. We have no figures for how often the crossing goes badly and are not inventing them.

There is a way around the assembly bill, and it changes the top end of what a clear is worth to one player. A character with enough level advantage can pull an instance apart one mob at a time, dragging single packs back down a corridor instead of fighting them where they stand, and clear the place without a group at all. The run takes longer than a full party would need, but nothing in it is shared: the coin, the cloth, the skins and any bind-on-equip drop all land in one set of bags, with no roll and no argument about who called it first.

That is why solo clearing reads differently on paper than it feels in play. A group splits 4 lines of value several ways and spends time forming first; a single player keeps all 4 and starts the moment the instance door is in reach. Against that sits the slower pace, the risk of losing the run to one bad pull, and the entry limit, which counts the same whether 1 player walks in or 5. Pulling one at a time is also the part that does not scale down well: an instance that resists it stays a group job. We have no list of which instances can be taken apart this way and at what level, and are not inventing one.

There is a thinner version of the solo run that skips the fighting altogether. Some instances hold chests that sit away from the packs, and a player who can reach them without being stopped takes the contents and leaves. Rogues have the clearest route to this, since the Eye of Bhossca chain ends in a disguise that lets a character move past mobs that would otherwise pull, which turns a chest route into a walk rather than a clear. Other classes have their own ways of covering ground quietly, but the disguise is the one that comes out of a quest chain rather than out of a cooldown.

What this does to the arithmetic above is change the unit being counted. A chest route is not a clear: it produces whatever the chest holds and nothing from the corpses, so the coin, the cloth and the skins all drop out of the calculation and only the chest line remains. In exchange the time per entry collapses, because the minutes that would have gone into pulling packs one at a time are not spent. That makes the account entry limit the binding constraint much sooner, since a route measured in a few minutes burns an entry just as completely as a full clear does. We have no list of which instances carry chests worth the walk, what they hold, or how often they are already looted when a player arrives, and are not inventing figures for any of it.

There is a ceiling above both of those lines. Testers report a limit on how many new instances an account may enter in an hour, named on the beta as 5, and it applies to the account rather than to the character. That turns clear speed into something with a hard stop: once the 5 entries are spent, the next run waits whether the group is assembled or not, and switching to an alt does not reset anything. A group that clears a short instance in well under 12 minutes does not earn proportionally more per hour, it simply reaches the cap sooner and then stands around.

The levelling argument runs the other way. Dungeon experience in Forever sits below what older notes assume, and the increase that followed the first round of complaints was small enough that most of the people making the argument kept making it. The common position is that a clear is a slower road to the next level than questing in the open world once the assembly time and the travel time above are counted against it. We have no figures for the size of the cut or of the increase that followed, and are not inventing them.

That does not settle anything for this page, because experience and gold are separate ledgers. A run that is poor value per level can still be the best hour available for cloth, skins, recipes and a bind-on-equip drop, and the materials line does not care what the experience bar did on the way. The practical read is to decide which of the 2 you are buying before the group forms. If the evening is about levels, the open world is the comparison to beat. If it is about the bags, put the clear against an hour of gathering using the travel and assembly times you measured yourself, and treat the entry limit as the thing that caps the night rather than the clear speed.

That cap is what makes the route argument worth having at all. Groups on the beta settle quickly on a preferred order through an instance: which packs get pulled, which side rooms get skipped, which bosses are taken because they sit on the way out rather than because anyone needs the loot. A settled route cuts minutes off a clear, and those minutes come straight off the denominator when the run is measured as gold per hour. The catch is that the saving only counts while entries are still available. Below the limit a tighter route means more clears in the same evening; once the hour's entries are spent, a faster route only buys standing-around time, and the group that took the long way through the optional wing is holding the same number of bags.

The instances named most often in beta reports are the ones with an obvious line through them. Deadmines runs close to linear, so the route question there is mostly about which optional packs get left standing. Upper Blackrock Spire is the opposite case: the approach alone has changed, the old shortcut is gone, and the order bosses are taken decides whether the group doubles back or not. Either way the number that matters is the one the group measures for itself, because a route that suits a party with a skinner and a tailor is not the route that suits one built for speed. We have no measured clear times for any instance on the beta and are not inventing them, and the same caution applies to anything that claims a fixed gold per hour from a named run. What the route does change reliably is how much of the evening is spent moving, which belongs in the same column as the standing bills of levelling rather than in the loot column.

The 9 starting dungeons and their brackets

The announced list covers the whole levelling grid, which is what makes per-bracket arithmetic possible at all. A run is only worth comparing against another run in the same band, since the cloth, the vendor trash and the bind-on-equip drops all move with level:

  • Hall of Thanes — 13-18, beneath Ironforge.
  • Ruins of Lordaeron — 15-20.
  • Excavation Site — 24-29, in the Wetlands.
  • City of Dalaran — 28-33, in the Alterac Mountains.
  • The Drowned City — 35-40, off Stranglethorn Vale.
  • Krol'dok Stronghold — 40-45, in Riverglades.
  • Alcaz Prison — 48-53.
  • Blackmaw Hold — 55-60.
  • Shaper's Terrace — 58-60.

Where no location is given above, none has been published that we can confirm, and we are not guessing at one. The brackets overlap by design, so most characters have 2 instances open at once and can pick the shorter approach rather than the higher level range.

The first 2 entries on that list carry a cost the others do not. The entrance to each lies inside a capital belonging to the opposing faction, so for half the player base the approach is not a walk to a door but a trip through hostile ground, with the corpse run and the repair bill that implies. Budget the travel separately from the clear: the instance itself may be short, while getting to it repeatedly across an evening is what eats the margin. We have no figures for how long that approach takes and are not inventing them.

The practical effect is on how income from dungeons gets estimated. Multiplying the contents of one clear by the number of clears that fit into the time available overstates the top end, because the entries run out before the hours do. The honest version counts entries first and time second: what an account can enter in an hour, what each of those entries reliably produces, and only then whether the group is fast enough to use all of them. Faster clears still matter, but mostly because they leave room for assembly and travel inside the same window rather than because they raise the ceiling. We do not know whether the figure stays at 5 outside the beta and are not assuming it does.

The other thing that stretches the hour is the repeat. Quest items that drop off a named boss do not share: if 3 members of the group need the same object, the object has to drop 3 times, which means the same corridor gets walked again on a night that was planned as one clear. Set pieces work the same way from the other direction, since the piece a player wants comes off one boss and nothing else in the instance substitutes for it. Neither case is a decision about whether the run was worth doing. The second and third entries are already committed the moment the group agrees to finish what it came for.

The instances that are not in the old list

Forum notes on the dungeon lineup include a set of instances with no counterpart in the original game, each with its own minimum entry level. They matter for income because they sit in level bands where the classic list is thin, which is where most levelling characters run out of places to go:

  • Excavation Site — Wetlands, 24–29.
  • City of Dalaran — 28–33.
  • The Drowned City — 35–40.
  • Krol'dok Stronghold — Riverglades, 40–45.
  • Alcaz Prison — 48–53.
  • Blackmaw Hold — 55–60.
  • Shaper's Terrace — Un'Goro, 58–60.

The entry level is the only hard figure here. What each of these drops, how long a clear takes and whether any of them can be pulled apart by a single character are all unknown to us, and the usual assumption that an old route guide still applies does not hold for an instance that did not exist before. Treat the bands above as a map of where the options are rather than as a ranking: the band a character is standing in decides which doors are open, and the 4 lines of value have to be measured inside each one separately.

Because the bind-on-equip drop is the only line that converts at market rate, it is also the only line that needs a rule agreed before the first pull. Three conventions are in common use: the item goes to whoever can wear it and nobody else rolls; everybody rolls and the winner does what they like with it, including selling; or everybody rolls, the winner lists the item and the proceeds are split when it sells. The third version is the one that produces arguments, because the roll happens in a minute and the sale happens later, sometimes several listings later, and the player holding the item is the one who fronts the deposit and answers the questions in whispers afterwards. State which of the 3 applies while the group is still at the door, not when the item is already in a bag.

For counting the drop towards what the run was worth, two habits are worth keeping. First, count what cleared, not what was listed: the number that belongs in the evening's total is the coin that arrived after the cut, minus any deposits spent on listings that did not sell. Second, do not average a bind-on-equip into every clear. One item at a good price can outweigh the coin, cloth and skins from a whole night, which makes it look like the run pays well on average, when in practice most clears produce none and the occasional one produces all of it. A group that budgets on the assumption that each run carries a sellable drop will be short most evenings. We have no drop rates for these items and are not estimating them, so the only honest version is to treat the coin and materials as the income the run can be relied on for, and the resale as the part that sometimes happens.

That is the version of the number worth having. A first clear looks good on paper because everything in the bags is new, while the repeat is where the real rate sits: mostly cloth, skins and coin, plus one more roll at the item that caused the return. Counting gold per hour on the repeat rather than on the opening run gives a figure that survives contact with an actual evening, and it lines up with the entry ceiling, because forced returns spend entries that cannot then be used on a different instance. The fair way to set it out is per entry rather than per clear: what the run reliably produces when the novelty is gone, multiplied by the entries the account still has left. We have no drop rates for quest objects or set pieces and are not guessing at how many returns a given item takes.

That list used to come second to the reason people walked in. Experience from the mobs inside a dungeon has been reduced, and the additional experience on dungeon quests was cut by 50% on top of that, so the bar at the bottom of the screen is no longer what an evening is planned around. What is left of a clear is the 4 lines above, and every one of them is paid in items or coin rather than in levels.

Players put the overall reduction at around 76% of the old rate. That number comes from the community rather than from anything published, so treat it as an opinion about the size of the change rather than as a measured figure. The direction is the part that holds: once the bar at the bottom of the screen stops being the reason to walk in, a clear is judged on what the bags hold at the end of it.

That shift changes what a dungeon is being compared against. It used to sit opposite the question of how fast a character reaches the next level, and the loot was treated as a bonus on top. Now it sits opposite a gathering route: both are ways of filling bags for an hour, and both are judged on what comes out at the end rather than on what the experience bar does. A clear that produces cloth, a recipe and one bind-on-equip drop is a supply run that happens to take place indoors.

Repeat visits follow from the same logic. Going back into the same instance a second and third time is now about a specific item on the drop table, not about the levels the clears would have carried before. That makes the decision narrower and easier to check: name the item you are there for, decide how many attempts you are willing to spend on it, and stop when the entries or the evening run out. If nothing on the table is worth the trip, the run has no second reason to exist, which is the practical difference between the current dungeon and the one older notes describe.

The cut lands unevenly across the clears of one evening. Quest rewards attached to an instance are collected once per character: the turn-in coin, the item from the reward list and whatever experience survives the 50% reduction all arrive on the first visit and never again. Mob experience, which used to carry the second and third run, is the line that was reduced hardest. So the first clear of an instance on a given character is a different proposition from the ones after it, and an estimate built on a single run repeated 3 or 4 times counts the quest side of the first visit several times over.

What is left for repeat runs is the part that does not expire: cloth, skins, vendor coin, recipes off the bosses and whatever bind-on-equip drops the tables give up. That is a real income line, but it is an item income line, and it competes with gathering on the same clock rather than with levelling. The practical reading is to treat the first run through an instance as a quest run and everything after it as a farm run, and to judge the farm runs on bag contents against the repair and reagent bill they generate. We have no figures for how much experience the mob reduction removed per clear and are not inventing any.

The obvious response, and the one raised every time the vendor-trash problem comes up, is that dungeons should simply drop more: fuller tables, higher cloth rates, bind-on-equip items on more bosses. The case for it is easy to state. A clear that ends in the red after a wipe would stop ending in the red, the assembly time would be covered by the contents of the bags, and the gap between an hour of gathering and an hour of group play would close without anything else being touched.

The counterargument is that most of what a dungeon produces is not unique to dungeons. Cloth, skins and the common crafting reagents are the same goods that gatherers bring in from the open world, sold on the same auction listings and bought by the same crafters. Raising the drop rate adds supply to a market that already has a supply side, and the price per stack moves down to meet it. The run gets fuller bags and a lower figure per item, and the player who spent the hour gathering instead takes the loss without having been in the instance at all. Recipes and bind-on-equip gear behave differently, because they are scarce by design and a buyer only needs one, but those are the lines that cannot be made generous without making the scarcity itself pointless. What profitability a clear currently has rests on the drop being thin enough that the materials hold their value, which is an uncomfortable position to defend and still the reason the numbers work. Whether any of these rates change after the beta we do not know.

The behaviour inside the instance changed with it. When experience carried the run, a group would finish a clear it was losing money on because the levels arrived anyway. Judged on loot output alone, a clear that drops nothing wearable in the first half has no reason to continue, and groups now break up part way through rather than walk the rest of the corridors for vendor trash. That shows up as players leaving after a named boss, after the one bind-on-equip chance most of them came for, or as soon as a wipe adds a repair bill to a run that was already thin.

Repeat clears of the same instance sit in the same bind. The second run through a place costs the same travel, the same assembly and the same repairs as the first, and it draws from the same drop tables, so the expected return does not improve on the back of knowing the pulls. Faster clearing shortens the evening rather than raising what the evening pays, and the entry limit caps how many attempts an account gets at the one line that converts into gold. Planning a session around 2 or 3 runs of a familiar instance now has to assume most of them end with materials and coin and nothing else, with the sellable drop treated as the exception rather than the budget.

Blizzard have described the reasoning behind the shift rather than just the numbers. The stated position is that quests were handing out too much experience and that mobs were handing out too little, and that clearing the same instance over and over was never meant to be the standard route through the levelling range. The rebalance is presented as moving experience back towards the open world, with dungeon spam treated as something the curve should not reward more than questing does.

For this page the useful part is who that leaves in the instance and why. If a clear is no longer the fastest way to gain a level, the players who keep running them are there for the 4 lines of value above: cloth and skins, the recipes off the bosses, a bind-on-equip drop that can be sold to a stranger, and the coin that mostly pays the repair. That changes the comparison with a gathering route as well, because the levels that used to be counted as part of the return on a dungeon evening are now largely on the questing side of the ledger. We have no figures for how the revised curve lands in practice and are not estimating any.

The reduction is being argued over in several threads at once, and the stated reason for it is worth repeating plainly: the cut is described as aimed at degenerate play loops, meaning the pattern where a character is parked in one instance and dragged through it on repeat instead of moving through the world. The common objection is that the loop being targeted was already the slower option. Players following a written questing route report levelling faster than a group chain-running the same instance, so the argument runs that the nerf removes a choice that was behind on speed to begin with rather than one that was ahead.

Whichever side of that is right, the effect on this page is the same. Once a clear stops being the fast way to a level, the people inside it change: fewer characters levelling through, more characters who are there for cloth, recipes or a specific bind-on-equip drop, and more who are being paid or carried rather than keeping pace with the group. That shifts who is selling runs and who is buying them, and it makes the 4 lines of value above the whole case for walking in rather than a bonus attached to the experience bar. It is also why the social weight on a single sellable drop goes up rather than down.

The shift also changes what a group is formed around. With kill experience reduced and the quest bonus cut by 50%, the reason to assemble is the quest reward and the drop table rather than the levels, and the two do not reward the same behaviour. Experience favoured killing everything on the way; quests and loot favour reaching the named mobs, the chests and the turn-in, and leaving the rest of the corridor alone. Groups that still clear wall to wall out of habit pay the extra time and the extra repairs for value that is no longer there, and the entry limit means that time cannot be made back by simply running the place again.

One consequence is already being discussed on the forums rather than observed in play: players expect level 60 characters to start selling quest runs, taking a lower-level group through the instance for the turn-ins and charging for it. The logic is that the quest line keeps its value after the experience nerf while the clear itself gets slower for anyone levelling through it, which is exactly the gap a paid service fills. Whether that market actually forms, and at what price, is unknown, and nothing of the sort is confirmed. It is worth tracking for one reason: if it does appear, the same instances start producing a second income line that has nothing to do with the drop table.

That reframing matters most for the instances where one faction has no quest chain at all. Shadowfang Keep sits in Horde territory and carries nothing for an Alliance character to hand in; the Stockade sits under Stormwind and offers the same blank to a Horde group. With dungeon quest experience cut and mob experience down with it, there was never much of a levelling case for walking into either of them from the wrong side, and now there is none worth planning an evening around. What remains is the drop table: a specific weapon or piece of armour wanted for an alt, cloth, skins, and whatever bind-on-equip the run produces.

Treat those runs as shopping trips and the time cost reads differently. The journey is the expensive part, because a character at the level the instance is built for is crossing contested ground to reach a door on the other faction's doorstep, usually on foot, and often more than once if the item does not drop the first time. Count the approach both ways before deciding, not just the clear, and decide in advance how many attempts the item is worth. If the same piece turns up on the auction house, buying it is frequently the cheaper answer, since the run costs an evening and the listing costs coin that an hour of gathering replaces. We have no drop rates for the items players usually make these trips for and are not inventing any.

The list of places this arithmetic applies to is longer than it used to be. Forever ships with 9 new dungeons at launch, and 3 of them were shown in Blizzard's video. Hall of Thanes sits under the throne room in Ironforge and fills the slot Ragefire Chasm fills for the Horde, which is the one already covered further down for the 13-18 band. Ruins of Lordaeron is undead and runs entirely in the open air, with no instance interior in the usual sense. Excavation Site: Wetlands is a gnomish dig, and the footage shows raptors sitting in tall grass rather than patrolling in view.

Each of those changes the shape of a run before any loot is counted. An open-air clear has no corridor to drag packs back down, which matters to anyone planning to take the place apart one mob at a time. A dig site with hidden spawns costs pulls that were not in the plan, and a pull that was not planned is the one that ends the evening. The entrance under the Ironforge throne is short on travel time for an Alliance character already in the city, which is a real part of what a clear is worth per hour. We have no drop tables, level ranges or clear times for the other 6 dungeons and are not guessing at them.

A later adjustment changed the shape of that reduction rather than reversing it. Dungeon quest experience is now paid on the first completion only, and experience from the mobs inside was raised by 20%. The 2 changes pull in different directions depending on who is walking in: a character entering an instance for the first time still collects the quest rewards attached to it, while a character going back in for a fourth clear gets nothing from the quest log at all and only the mob line remains.

That moves the repeat run out of the levelling column and into the earning one. Every clear after the first is judged on drops, materials and vendor coin, with whatever experience the mobs pay as a side effect rather than the reason to be there. It also makes the comparison with gathering an honest one: both are now being measured on what the bags hold at the end of the hour, instead of one of them being credited twice for paying in levels as well as in items.

That shifts the comparison. A dungeon used to be weighed against questing, where the question was which one levelled faster; now it is weighed against gathering, where the question is which one produces more sellable goods per hour of play. Assembly time, travel on the current geometry and the entry limit all count against the run in that comparison, and none of them had much weight when levels were the payout. Anyone still budgeting dungeon nights the old way is measuring against a reward that is no longer the main one, and the rest of the levelling economy in Forever has to absorb the difference.

One thing to settle before planning a route around a favourite instance: there is no heroic or scaled version of the low-level dungeons, so the places that pay well while you are climbing stop paying once you are at the top. Hall of Thanes and Razorfen Kraul are the usual examples. Both are reasonable stops on the way up, both have drop tables fixed at the level they were built for, and neither has a harder mode that would put current materials or current recipes behind the same bosses. A character at cap who walks back in gets the same cloth, the same vendor trash and the same bind-on-equip items aimed at levellers, with the clear taking less time but producing nothing a cap character needs.

That matters for how a money route is written down. A list of instances built while levelling quietly expires at the top, and the entries it spends are the same 5 an hour an account gets for anything else, so re-running an old favourite costs the same ceiling as a run that still pays. The items from those clears remain worth something to the people still levelling, which is a supply argument rather than a farming one: you are producing stock for a buyer, not gear for yourself. Whether any of the low-level instances later get a version tuned for cap characters is not something we have information on, and we are not assuming it either way.

Quest rewards moved too, and not in your favour. The rewards handed out by the quests tied to the Excavation Site were reduced to uncommon quality, which takes a chunk out of the one line on the list that used to be dependable. A quest reward is the part of a run you can count on before you walk in: the slot is covered whether or not the drop table cooperates. When that reward comes back at uncommon rather than better, the slot is still covered, but the item is closer to what a stranger would sell you on the auction house than to something worth planning the evening around.

The practical effect is to push the value of an Excavation Site run further towards the other 3 lines. Mob kills, cloth and skins, and whatever bind-on-equip piece turns up are now carrying more of the clear, because the guaranteed reward at the end carries less of it. That makes the instance a weaker pick for a character short in a specific slot and a slightly better one for a group that is there to fill bags, since the material and drop side of the run was not touched. We have no figures for what the rewards were before the change and are not inventing them.

The reason players give for that cut, in the threads where it comes up, has nothing to do with loot. It is described as a measure against boosting and against characters running the same instance on repeat for levels, since the bar at the bottom of the screen was the thing both of those were built on. Drop tables, quest rewards and the vendor value of what fills the bags are not part of that argument, which is why the income side of a clear reads the same after the change as before it. We have not seen an official statement setting out the reasoning, so this is the explanation players are circulating rather than a confirmed one.

The practical consequence is the frame the rest of this guide uses. A dungeon is now judged on money and loot: slots covered by quest rewards, patterns that come off bosses, and the bind-on-equip pieces a stranger will pay for. Whether an evening in there was worth it is answered by the bags at the end of it and by how much of the standing bills of levelling they cover, not by how far the bar moved. We do not have figures for how far either experience cut goes and are not inventing them.

That reframing changes which instance you pick, not only whether you go in at all. When the bar was the point, the choice between 2 dungeons open at the same level came down to which one cleared faster per level. With the experience cut in place, the comparison is between drop tables and quest chains instead. The instance worth the evening is the one with quest rewards for the slots you are actually short in, a pattern you intend to use or sell, and a bind-on-equip list a stranger would pay for. Two instances that move the bar at the same rate can read very differently on that test.

The number of instances that test is applied to grew before launch. The original announcement put 4 new dungeons in the game; the figure stated for launch is 9. That is more than twice as many fresh drop tables, quest chains and boss pattern lists to compare against each other, and it lands on a frame where the comparison is already about loot rather than levels. A wider menu at the same level bracket means fewer evenings spent on an instance picked because it was the only one open, and more spent on the one holding the pattern or the slot you are short in.

One of those instances is already open to testers. City of Dalaran is on the beta, entered through the sewers rather than through a door on the surface, and Horde characters have to finish a quest chain before the instance will let them in. Alliance access and Horde access are therefore not the same errand, which matters when a group is mixed on a test realm and half of it is still working through the chain.

There is also a boss inside that does not appear on the normal route. It is summoned with an item, and that item carries a 3-minute cooldown, so a group that wants the extra kill has to plan the pull rather than stumble into it. What the drop table behind that boss looks like, and whether anything off it is bind-on-equip, we do not know yet: the instance is on the beta and we have no loot figures for it. It belongs on the list because it is being tested, not because we can price a run through it.

The comparison players argue about most on the forums is dungeon against quest, and it is usually set up as a time problem. The claim is that a run of around 20 minutes hands out loot on the level of a long quest chain and moves the bar as well, while a chain of the same length pays once, at the end, in a single reward item. Nobody in those threads has posted a measured set of numbers for either side, so the figure to treat as given is the run length people quote rather than the loot comparison built on top of it.

What is worth noting is that the experience cut and the income question are argued in separate threads. The complaints about reduced mob experience and the 50% cut on dungeon quests are about levelling speed and about boosting; they do not touch drop tables, cloth, patterns or bind-on-equip pieces. Taken only on money, a clear still compares well against a quest chain of similar length, because the lines that pay were left alone. That is the frame to use when deciding whether an evening covers the standing bills of levelling: the bar moves less than it did, the bags do not.

It also thins out the case for running the same place on repeat. With 4 destinations, a group short on options clears what it has; with 9, the second evening has somewhere else to go, and the bind-on-equip lists are not the same across all of them. What we do not know is which of the 9 sit at which levels, how many are genuinely new versus reworked, or how their drop tables compare, so there is no ranking to give here. The practical advice stays the same as above: judge each one on the slots it covers and the items a stranger will pay for, and count the clear against the standing bills of levelling rather than against the bar.

The same logic applies to repeats and to travel. Running one instance back to back stopped being a levelling plan, so a second clear has to justify itself on a specific item: the pattern that did not drop the first time, the slot a quest reward covers, the bind-on-equip piece you meant to list. Travel out to a distant instance used to be partly paid for by the bar moving on the way through, and now it is cost against the bags and nothing else, so a closer dungeon with a thinner drop table can still come out ahead on an evening meant to cover part of the standing bills of levelling. We have no figures for travel times or clear times and are not inventing them.

The material line is the one that does not care about the bar. Cloth drops at the same rate whether the experience from mobs inside is full or cut, and so do the skins off whatever the group is willing to stop for. That makes a run planned around bag space a different activity from a levelling clear: the number worth watching is humanoids killed per hour, not bosses downed, and a loop that comes back through respawned trash can be worth more than a clean path to the last boss and out. Groups built around a character who can hold several packs at once have always been the efficient version of that, because cloth is paid per corpse rather than per clear.

Two things sit on top of that plan and neither is settled. The reduction to dungeon mob experience and the 50% cut to dungeon quest experience remove the side benefit that used to make a cloth route worth doing even on a bad drop night, so the evening now has to pay for itself in bags alone. And anything that changes how safely one character gathers and holds a large pull - kiting, aggro behaviour, the tools used to bring trash together before it dies - changes how many corpses an hour produces, which is the only input a material route runs on. We have no figures for either and are not inventing them, so a route is something to re-time yourself after a patch rather than a fixed rate. It is still the steadiest way to turn an evening indoors into coin towards the riding bill at 40.

Worth stating plainly, because it is the part that catches people out on the second evening in the same place: the quest side of a dungeon pays once. The reward slots, the coin attached to the turn-in and the item handed over at the end are all consumed the first time the chain is finished, and they do not come back for that character. From the second clear onward the instance pays only on the lines that reset, which is mob coin, cloth and skins, boss patterns and whatever bind-on-equip pieces drop. That is a real step down in what a run is worth, and it happens without anything inside the dungeon changing.

This is why repeatable turn-ins come up whenever the subject is discussed. A quest that can be handed in again and that pays in coin rather than in an item would put a floor under repeat clears, since the group would know what the evening pays before walking in rather than waiting on a drop table. Players trade notes on which turn-ins behave that way and which ones are one-time only, and the answers are not consistent enough for us to repeat as fact. We have not confirmed which repeatable quests exist in dungeons on Forever, how often they can be handed in, or what they pay, so treat a repeat run as paying on drops and materials alone until you have checked the specific instance yourself. If the plan behind the evening is the riding bill at 40, that assumption is the safer one to budget against.

Since the experience cuts landed, the shape of the argument has changed. It used to be whether a clear beat questing for levels; now the threads are about whether a run comes out roughly level with efficient questing on speed, and several posters put their own figures on both sides to make the case. Those figures are claims by the people who posted them, drawn from their own evenings, and they are worth reading as that rather than as measured rates. No official numbers have been published for either side of that comparison, so we are not treating the forum totals as data.

On money the two do not line up the way the speed argument suggests. A gathering route pays in a steady stream of the same few materials, can be stopped at any point, and needs nobody else to show up. A dungeon pays in cloth, skins, the occasional pattern and whatever bind-on-equip piece turns up, which is lumpy and depends on a group forming first. Two evenings can finish at similar totals and still feel nothing alike, because one chips away at the riding bill at 40 hour after hour while the other either covers a stretch of it in one drop or returns almost nothing. Settling the question on speed alone answers something other than what your bags are asking.

So the rest of this guide is a loot and gold calculation rather than a time one. The question is not whether a run levels faster than questing, it is whether what ends up in the bags covers more of the standing bills of levelling than the same hour spent on a gathering route would. That is a harder comparison to win, because a gathering hour produces something sellable every time and a clear only does when the right thing drops. We do not have figures for how far either cut goes and are not inventing them.

On 8 October that calculation moved again, in the other direction. Experience from kills inside a dungeon was raised by 20%, and a bug was fixed where the experience a kill handed out was calculated from the level of the highest-level character in the group rather than from your own. The second part matters more than the first for anyone who levels with a friend several levels ahead: before the fix, the person at the bottom of the group was being measured against somebody else's level. Neither change touches drop tables, quest rewards or vendor value, so the 4 lines of income above read exactly as they did.

Dalaran is the clearest example of a dungeon that pays in access rather than in coin. The instance is entered through the sewers, and a full clear takes the city's reputation to just short of friendly, which is the point of going in for most groups. What that reputation opens is modest: the open part of the city holds a quartermaster with a couple of recipes, 1 or 2 quest-giving NPCs, and no full shops behind them. There is no auction house, no bank row, no set of trade vendors to work against.

Treat it as a supply line with 2 or 3 entries on it, not as a trading hub. The recipes are the reason to finish the clear, and anything else the evening produces still has to be carried back to a city with an auction house before it turns into gold. For an evening meant to cover part of the standing bills of levelling, the Dalaran run is worth planning around the patterns and the quest rewards, with the usual loot lines on top. Whether more vendors open behind higher standing is not something we can state, so we are not guessing at it.

What it does change is the honest version of the comparison. A clear is no longer a pure loot errand with the bar standing still, so the choice between a dungeon run for loot and quests and an hour on a gathering route now comes down to what you need next. If the gap is gear, the dungeon covers slots you would otherwise buy and pays part of the standing bills of levelling in kind. If the gap is coin for the riding bill at 40, the gathering route still wins on consistency, because it produces something sellable every time out and a clear only does when the right item drops. We have no figures for how the 20% compares with the earlier reduction and are not estimating them.

City of Dalaran opened for testing in week 4 of the beta, which adds another instance to the list without yet telling us what it is worth. We have no drop tables, no quest rewards and no vendor figures for it, so it sits outside the 4 lines of income above until players run it enough to report back. For now it is a dungeon to look at rather than one to plan an earning evening around.

When numbers do appear, the test is the same one the rest of this guide uses: how many gear slots the quest rewards cover, whether any bind-on-equip piece off the bosses is something a stranger will pay for, and how much of the standing bills of levelling the bags cover on the way out. Until then we are not putting a figure on it.

Access has since been described in more detail. City of Dalaran is open on the beta, and Horde characters reach it after a questline that ends with a key to the sewers, so getting in is itself a job of work rather than a walk to the entrance. That does not change the 4 lines of income above, but it does mean the first clear costs an evening of questing before any loot appears, and that time has to be counted against whatever the bags hold at the end of it.

Two other adjustments land on the same comparison. The quest rewards from Excavation Site: Wetlands were lowered in rarity, which takes a slice out of the gear-slot side of a run there: the slots are still covered, but with weaker items, and fewer of them are worth holding rather than vendoring. Against that, experience from mobs inside dungeons is up by 20%, so the levelling case for walking in is stronger than it was while the loot case for that particular instance is weaker. Which way the trade falls depends on whether the gap is gear or coin for the riding bill at 40. We have no replacement drop tables or reward lists for either change and are not inventing them.

The argument about Need

There is a proposal on the official Forever forum to make a bind-on-equip item soulbound the moment somebody rolls Need on it. The reasoning is straightforward: rolling Need is supposed to mean I will wear this, and at the moment it can also mean I will sell this, with no way for the group to tell which.

The reply that gives the thread its shape is not about greed at all: need for gold to buy next abilities.Player, official Forever forum

That is the whole economy of levelling in 6 words. The person rolling Need on a sellable item is often not a profiteer; they are somebody who cannot afford their next rank of a spell. Both sides of the argument are describing the same shortage from opposite ends.

Nothing about loot rules has been announced. This is a request players are making, and we are reporting it as one.

The bill nobody budgets for

Abilities cost money. Every few levels the class trainer wants paying for the next rank, and that expense arrives on a schedule regardless of whether the last 3 dungeons dropped anything worth selling. It is the most reliable drain on a levelling character and the least discussed.

It is also why the standard advice to "just save for riding" is incomplete. Riding is the large bill at 40, but the training bill is the steady one before it, and the two compete for the same coin. A character that spends everything on spell ranks arrives at 40 with nothing banked; a character that banks everything arrives with a mount and yesterday's abilities.

We do not have Forever's trainer prices and we are not going to invent them. What is worth knowing now is the shape: this is a recurring cost with no upper bound in sight, and it is the reason a group of 5 levelling players will argue over 1 sellable drop.

Hall of Thanes, and why the quests matter more than the coin

Hall of Thanes sits under the High Seat in Old Ironforge, and it is better read as a gear source than a coin source. 4 bosses hold it: Faldrim Oldanvil, Magmatus, Krad and Dargen Crushhammer. Attached to the run are 5 quests, and the chain starts outside the instance with Beldin Steelgrill at Steelgrill Depot. Those quests are Alliance only, which is the detail that changes the arithmetic depending on which side you play.

The 5 by name, so you can check them off as you go:

  • Important Heirlooms
  • The Restless Dead
  • Old Ironforge Incursion
  • An Ancient Grudge
  • The Treaty of Understanding

All 5 are Alliance only. That is not a flavour restriction, it is a line in the budget: an Alliance character running the place works through the quest log and comes out with gear, while a Horde character clears the same 4 bosses and then buys the equivalent slots. We do not have the order the 5 unlock in beyond the start with Beldin Steelgrill, and we are not going to guess at it. What the list is useful for is planning, because coin saved on gear is coin left for the standing bills of levelling, and the training bill above is the first one to arrive.

The rewards are the part that touches your purse, because each one is a slot you do not have to buy:

  • Ironforge Greathammer
  • Deepblaze
  • Dwarven Tome
  • Catacomb Cloak

For an Alliance character that is 4 slots covered by questing, which is 4 fewer orders placed with a crafter or bid on at the auction house. A Horde character clearing the same dungeon fills those slots the usual way and pays for them, so the faction restriction is a real cost difference rather than a flavour note. It also feeds back into the crafting side, since demand for the same pieces comes mostly from the players who cannot quest for them, and that is worth keeping in mind when working out which profession pays its way. We do not have stats or item levels for these rewards and we are not guessing at them; the point is how many slots they cover.

Ruins of Lordaeron works the other way round. It sits under Undercity, it is built for levels 16 to 22, and in tone and location it is Horde ground, but the quests attached to it are not restricted to one side: both factions have quests here. That matters for the same reason the Hall of Thanes restriction matters, only in reverse — an Alliance character is not locked out of the reward slots, so the gear you can quest for is coin you do not spend at the auction house regardless of which side you play. We do not have the boss list, the quest names or the rewards for this one, and we are not inventing them.

With the ceiling at 30, the list of instances worth costing out runs further up than this guide originally covered. Excavation Site: Wetlands is built for levels 26 to 31, Razorfen Downs opens from 25, and Uldaman from 30, which puts the last of them right against the cap. The practical consequence is that a character near the ceiling has 3 more places to clear and no levels left to gain from them, so the whole reason to walk in is the gear, the recipes and the bind-on-equip items. We do not have boss lists, quest names or reward tables for these 3 and are not inventing them.

City of Dalaran belongs on the same list, opening at level 30, and it is the one instance where the cost of a clear differs before anyone has pulled anything. An Alliance character collects the key from the Image of Archmage Modera and can be at the entrance the same evening. A Horde character does not get a key handed over at all: the way in runs through a quest chain starting at Tarren Mill, and that chain has to be finished before the door opens.

The Horde steps, in the order they are given, are Prison Break In, Key to the City, Stone Tokens, Dalaran Patrols, Bracers of Binding, Blood in the Streets, Heart of Disruption and Dalaran Sewer Key. For budgeting that is an evening of open-world work priced separately from the clear itself, done once per character, and it is paid in travel and quest turn-ins rather than in coin. The practical effect is the Hall of Thanes restriction in another form: the 2 factions walk out of the same dungeon with the same loot, but one of them spent an extra session earning the right to walk in. If the coin is wanted by a fixed date, that session competes directly with the levelling income plan by level, and the bills behind it are the ones listed under the standing bills of levelling. We do not have the boss list, the reward tables or the order the Horde steps unlock in beyond the names above, and we are not inventing them.

Those 3 opened in the build of 1 October, and the same build trimmed both halves of the old reason to go in: experience from dungeon quests was reduced, and so was experience from the mobs inside. What was added on the other side of the ledger is loot rather than levels — blacksmithing and leatherworking recipes now turn up as random drops from dungeon content instead of sitting only with trainers and vendors. So the build moved the payout from the bar at the bottom of the screen to the bags, which is the same direction the rest of this guide describes, applied to 3 more instances at once.

The order of those 2 cuts is worth keeping straight, because it explains how the place is being talked about now. Experience from the mobs inside was trimmed first. The additional experience on dungeon quests was then cut by 50% on top of that, so the quest log stopped being the half of a run that held its value while the rest fell away. Excavation Site: Wetlands opened alongside the same set of changes, which means the newest instance on the list has never been available as a levelling route at all — the first groups through it were already judging it on drops, quest rewards and what a stranger will pay for the bind-on-equip pieces.

For planning purposes that makes the entry simpler to score than the older dungeons, since there is no levelling half to weigh against the loot half. An evening in there is a fixed clear for a fixed payout: slots covered by quest rewards, whatever recipes fall out of the bosses, and vendor coin from the rest of the bags after repairs. Set that against the levelling income plan by level rather than against another dungeon, because the comparison with a gathering route is the only one left. We do not have boss lists, quest names, reward tables or drop rates for this instance, and we are not inventing them.

What we do have for Excavation Site: Wetlands is the shape of the quest log, and it is uneven between the factions. The instance is built for levels 26 to 31. A Horde character has 4 quests attached to it; an Alliance character has 7. That difference works the same way the Hall of Thanes restriction does, only smaller: the side with more turn-ins covers more slots from the quest log and places fewer orders for the same slots at the auction house. We do not have the names of those quests, their stats or their item levels, and we are not inventing them.

The part worth planning around sits at the end of the clear. The last boss drops a Titan Relic, and the relic starts a chain rather than being a reward in itself. Finishing that chain offers a choice between a cloak and a weapon, so you take 1 and leave the other, and it also leads to a trinket that is available to both factions. The practical consequence is that the first visit is the expensive one to skip: the relic has to drop, the chain has to be worked through, and the cloak or weapon only lands if you see it to the end. Price a repeat clear as vendor coin plus whatever a stranger pays for the bind-on-equip pieces, and keep the standing bills of levelling in view when deciding whether that is worth an evening. We do not have drop rates for the relic or stats for any of the 3 rewards.

For budgeting, a random recipe is the awkward kind of income: it is worth real coin when it lands and it cannot be planned for, so it belongs in the upside of an evening rather than the forecast. The steadier part of the same change is that 2 crafting lines now have a supply route that does not run through a trainer, which is worth weighing before committing to blacksmithing or leatherworking in Forever. A character at the ceiling clearing Uldaman is there for exactly these 3 things: slots covered, patterns learned and bind-on-equip pieces a stranger will pay for. We do not have drop rates for the recipes, a list of which ones are in the pool, or figures for how far either experience cut goes, and we are not inventing them.

The other thing to price in at this end of the range is the quest experience. Dungeon quests have been reduced by a further 50% on top of the cut that already applies to the instances themselves, so the argument earlier on this guide — that the quest log is the one part of a run still paying experience — holds in direction but not in size. Finishing a chain is still worth doing for the blue at the end and for the slots it covers; it is no longer a sensible way to move the bar. If the evening is being spent on income rather than gear, an hour in Uldaman has to beat an hour on the levelling income plan by level on bag contents alone, because nothing else is coming out of it.

The clearest version of that comes from the beta threads themselves: once the chain is turned in, players there say outright that there is no reason to walk back in. That is sentiment from testing rather than a measurement, but it sits consistently with the 2 cuts described above. Quest rewards are collected once, the experience no longer moves the bar, and the clear that remains is scored on 2 lines only — the vendor value of what fills the bags, and whatever a stranger pays for the bind-on-equip pieces.

For budgeting that means a first visit and a repeat visit are worth different amounts, and the gap between them is wider than it used to be. Price the first clear with the slots covered and the chain finished; price every clear after it as vendor coin plus the chance of a saleable drop, minus repairs and the time spent forming a group. On those terms a repeat run competes with a gathering hour rather than with another instance, and it loses that comparison on any evening where nothing saleable lands. Keep the standing bills of levelling in view when making the call. We do not have drop rates for the saleable pieces and are not inventing them.

That one-off shape is what breaks the per-hour comparison most people start from. A gathering route can be run again tomorrow and the evening after that, and each pass pays roughly what the last one did. A dungeon whose chains are finished pays its full amount once per character, and every clear after that is the thinner version described above. So the honest unit for a run is not gold per hour but gold per character: the slots the quest log covers, the patterns that happened to drop, and whatever a stranger pays for the saleable pieces from the 1 evening it was worth forming a group for.

Budgeting from that unit changes 2 things. A new character has a one-off payout sitting in each instance it can reach, and it is worth collecting while the chains are still live rather than arriving at the ceiling with turn-ins left behind. A character that has already been through them has no dungeon income left to plan around and is back on the levelling income plan by level for anything that arrives on a schedule, which is most of the standing bills of levelling. We do not have figures for what a first clear leaves in the bags, and the total depends on drops we cannot predict.

The 2 cuts land on the same side of the ledger: quest experience in dungeons is down 50%, experience from the mobs inside is reduced as well, and what was added in exchange is loot. Blacksmithing and leatherworking recipes now turn up as random drops from dungeon content instead of sitting only with trainers and vendors, so a clear produces coin and patterns where it used to produce levels. Going in is no longer a levelling decision at all.

That changes the comparison the rest of this guide is built on. A gathering route and a dungeon clear used to be scored on different things — one paid in materials, the other in experience — and the dungeon won whenever levels were the bottleneck. With experience trimmed on both halves, both are judged on the same question: what is in the bags at the end. A run wins on slots covered, on a pattern a crafter would otherwise charge for, and on bind-on-equip pieces a stranger will pay for; it loses on any evening where none of those land, because a gathering hour always produces something sellable. Anyone weighing a crafting line against this new supply route should read it next to what Forever changed about professions. We do not have drop rates for the recipes, the list of which ones are in the pool, or a figure for the cut to experience from mobs, and we are not inventing any of them.

One practical note applies to both of the new dungeons: they do not come with a built-in map. The interface shows nothing for the layout, so the first clear is spent working out where the corridors go, and a group that takes a wrong turn pulls packs it did not intend to pull. That is a time and repair cost on the first visit rather than a permanent one, but it is worth planning for if the evening has a deadline attached to it. Run the place once to learn it, then decide whether the quest chains are worth coming back for.

One change works against that arithmetic. Dungeon quest rewards have been pulled down to green quality across the board, with blue kept for the quest that closes a chain. So the 4 items above are not 4 equal prizes: most of a chain now hands you an ordinary item, and the one piece worth planning around sits at the end. A character who runs the dungeon twice and abandons the chain halfway collects the cheap half of the reward and none of the expensive half.

For the dungeon-against-gathering comparison already on this guide, that shifts things by a slot or two rather than overturning them. A green reward saves you the price of a green, which is small, and it still covers a slot you would otherwise fill at the auction house. The blue at the end of a chain is the part that replaces a real purchase, and it only pays out if you finish. If you are picking a route for the evening, the honest version is: run the dungeon when you intend to see a chain through, and take the levelling income plan by level otherwise, because half a chain is closer to a gathering hour in value than it looks. We do not have a list of which quests count as the final one in each chain and we are not guessing at it.

The comparison players are drawing in response is with the crafting bench rather than with another instance. If the payout for 3 or 4 turn-ins is a green, a blacksmith or a leatherworker can cover the same slot without forming a group, without the travel and without waiting on the last player, and the materials for it come off a gathering route that produces something sellable every hour. That is the argument as it is being put: not that a green reward is worthless, but that the cheaper path to an ordinary item no longer runs through a group of 5.

It does not hold for every slot. The blue that closes a chain is the piece a crafter would charge real coin for, and the recipes off bosses exist nowhere else. The honest split is that green-tier rewards are usually matched more cheaply at the bench or at the auction house, while the closing reward and the patterns are what make the clear worth forming a group for. Anyone deciding which side of that line to build a character around should weigh it against which profession pays its way. We do not have a list of which quests were reduced to green quality, and we are not guessing at it.

The other change pushing in the same direction is the cut to experience from dungeons. When the run stopped being the fastest way to level, groups stopped treating it as something to repeat. What you see now is 1 clear, taken at pace, with trash packs pulled only when they stand between the group and the next boss. That is a different activity from the old chain of back-to-back runs, and it has to be valued differently.

That has a knock-on effect outside your own bags. If most groups walk in once per character, kill the bosses that stand in the way of a quest chain and leave, then the number of trash packs being killed across the server falls with them, and every bind-on-equip piece that used to come off those packs falls with it. Fewer saleable items reach the auction house from dungeons, and the ones that do arrive from a 1-off clear rather than from somebody farming the same corridor all evening. For a seller that means less competition on the listings and a thinner, less predictable supply; for a buyer it means the slot you wanted may simply not be listed on the night you look. Either way the dungeon is a smaller feed into the item market than it was, and the steady half of an evening's income has to come from somewhere else, which is the case the levelling income plan by level is built on. We do not have figures for how much of the saleable supply came from trash rather than bosses, and we are not inventing them.

There is one more instance worth noting for planning, with the caveat that almost nothing about it is known. Mistmantle Manor appears in the game files as a 5-player zone, which puts it in the same category as the dungeons costed out above rather than with the open-world content. We do not have a level range, a boss list, quest names, reward tables or a location for it, and a name in the files is not a release date. Treat it as a line item to check later rather than as something to budget an evening around, and keep the planning on the instances that are actually open and on the standing bills of levelling they are meant to cover.

The practical effect is that a dungeon evening is measured by what the bosses and the quest log hand over, not by how many times you go back in. Skipped packs mean less cloth, fewer skins and less vendor coin, so the reliable half of the income list shrinks. What survives the skipping is the part that only drops from bosses: bind-on-equip items, recipes and the blueprints that came in with the profession changes in Forever. Plan the run around those and the quest chain you intend to finish, and treat the trash as cost rather than income. We do not have figures for how far dungeon experience was reduced and we are not going to put a number on it.

With dungeon quest experience cut by 50% on top of the trim to the mobs inside, the turn-ins stopped being the half of the run that held its value. A clear is now scored the way a gathering hour is scored: slots covered by quest rewards, patterns learned, and what a stranger will pay for the bind-on-equip pieces. Levelling speed has dropped out of the comparison entirely, so the only sensible question before forming a group is what the evening leaves in the bags, not how far the bar moves by the end of it.

That shows up in how groups are put together. A corpse is skinned once, so a second skinner splits no extra supply and adds nothing to the takings, and the composition people settle on now carries 1 skinner with the other 4 slots filled on what the run needs rather than on who can loot it. The same applies to the cloth, which goes to whoever wants it and is thinner anyway once trash packs are being walked past. Materials are still the reliable line in the income list, but they are a smaller line than they were, which is worth holding against the standing bills of levelling before committing an evening to a run. We do not have figures for either experience cut and are not inventing them.

The thing that decides the shape of a run is not who tanks it but whether the group agreed to skip. With experience from the mobs inside devalued, a clear is planned as a boss-only route: pull the packs that stand in the corridor, walk past the rest, and stop when the last boss is down. Everyone joining a group is implicitly agreeing to that route before the first pull, which is why a player who wants a full clear for materials tends to be told so at the summoning stone rather than halfway in.

The lines that shrink are the dependable ones. Vendor coin, cloth and skins all scale with packs killed, so a skipped route removes most of the income that used to arrive whether or not a saleable item dropped, and leaves the evening resting on boss loot alone. The second cost is the one people forget to count: quest objectives sitting on trash are walked past as well, so collecting them means forming a group again, travelling again and paying a second repair bill for the same instance. Budget that as a separate evening rather than a detour, and compare it with the levelling income plan by level before deciding the quest items are worth the return trip.

It is worth being blunt about what the experience cut does to motivation. A dungeon used to pay twice: levels on the one hand, gear and coin on the other, and either half alone was enough to justify going back in. With the experience half reduced, the second run of the same instance in one evening has to stand on loot alone, and loot from a repeat clear is the same 4 bosses rolling against the same table. That is why groups disband after 1 clear now instead of resetting and queueing the next one.

Which leaves the instance as a 1-visit item on the schedule and the open world as the place the rest of the evening is spent. Humanoid packs outside drop the cloth that used to come off trash inside, they drop greens at the same unremarkable rate, and the coin that pays for the next stack of potions and bandages comes off the same route. None of that is exciting, but it arrives every hour rather than once per character, and it is what a levelling player is actually living on between clears.

So the honest split is: the dungeon covers slots, patterns and the occasional saleable piece, once. The grind outside covers the consumables and the running costs. A character that treats the run as its income plan ends up short the week the drops do not land, which is the case the levelling income plan by level is built around, and the bills it has to meet are the ones set out under the standing bills of levelling. We do not have drop rates for cloth or greens from open-world mobs and are not inventing them.

A forum breakdown puts rough figures on that gap, and they are worth quoting as the players' own arithmetic rather than as measured data. At level 17, the post reports about 15 experience from a mob inside a dungeon against 170 to 200 from a mob of comparable level outside it. Whatever the exact figures turn out to be across the range, the direction is the one every thread describes: the bar barely moves while you are in there, so the clear has to justify itself on what it drops.

Two things are being argued as one in those threads, and they are worth separating. The reduction is aimed at bots and at characters grinding instances alone, and what it touches is the rate at which the bar fills. Drop tables were not changed by it, quest rewards were not changed by it, and the vendor value of what fills the bags at the end of a clear is what it was before. A run that covered its own repairs and left a bind-on-equip piece over still does.

So dungeons were nerfed is accurate about levelling and misleading about income. If the evening is being planned for coin, the questions are the ones already set out above — slots covered by quest rewards, patterns learned, and what a stranger pays for the saleable pieces — and none of those answers moved. What the cut changes is the justification: the levelling half of it is gone, so the clear has to stand on the loot half alone, measured against the standing bills of levelling. We have not seen an official statement spelling out the reasoning, so treat the anti-bot explanation as the one players are circulating rather than as a confirmed one.

The defence of dungeons in those same threads takes that split as given. Nobody arguing the other side claims a clear is still a quick way to level. The case made for walking in is that the instance is where gear and saleable items come from, and that the bar at the bottom of the screen was never the only reason to form a group. Put that way the complaint and the defence are not really in conflict: one is describing the levelling column of the plan, the other the income column, and only the first of the 2 was touched.

For the choice in front of you that narrows the comparison rather than settling it. A dungeon is no longer weighed against questing on levelling speed, it is weighed against a gathering route on what the bags hold at the end — slots covered by quest rewards, patterns off bosses, and the bind-on-equip pieces a stranger will pay for. If the evening is about gear, the instance is the obvious place to spend it; if it is about coin arriving on a schedule, the levelling income plan by level is the steadier route, and the bills both are measured against sit under the standing bills of levelling. This is the position players are taking during testing rather than anything confirmed.

The same threads describe where those players go, and that complicates the comparison this guide rests on. With a clear no longer paying levels, a share of the groups that would have formed are out in the open world instead, and the open world has friction of its own: a mob belongs to whoever tags it first, so a contested camp can leave you waiting on respawns rather than killing anything. Players are raising that as a real problem rather than a background annoyance, which means the gathering or questing hour a dungeon is being measured against is not always the clean hour it looks like on paper.

Behind that sits the older argument about dungeons against quests as the levelling route, and it decides what is on the auction house as much as it decides how fast a character reaches the ceiling. A player who levels inside instances is not standing on a gathering route: the ore, herbs and leather along the way are never picked, and none of those hours produce a stack for somebody else to buy. A player who levels on quests in the open world passes nodes whether or not gathering is the plan, and a share of what they pick is listed as a side effect of the evening rather than as a decision.

So the 2 sides of that argument pull the same lever in opposite directions. If instances become the main route, fewer characters work the zones, less raw material reaches the auction house, and a gathering hour is worth more to whoever still does it while crafters pay more for the same stack. If questing is the main route, the zones are busy, the material listings are long, and the friction moves to competition for nodes and tags instead. Either way the levelling route a player picks is also a supply decision, which is worth weighing against which profession pays its way before committing a character to one. We do not have figures for how that split stands at present, or for how many players have moved either way, and we are not inventing them.

The practical version for planning is to score both sides on the evening in front of you rather than in the abstract. If the zones you work are busy, an hour outside can finish with less in the bags than it usually does, and a clear with a chain still to finish holds up better by comparison; if they are quiet, the open route wins on steadiness for the reasons already set out in the levelling income plan by level. Either way the question is what the hour leaves in the bags against the standing bills of levelling, and it is worth re-checking as players move between instances and the open world rather than deciding once and sticking to it. We do not have figures for how many players have shifted either way, and we are not inventing them.

That is the context in which the repeat clear is defended at all. The case players make for it is not that a second run pays well, it is that on a crowded evening the alternative is a route where the nodes have already been picked and the mobs are tagged before you reach them. Inside an instance the spawns belong to the group for the length of the clear, nothing on the path is taken from in front of you, and the drop table behaves the same no matter how many other groups are in the same dungeon elsewhere. The argument is about competition outside rather than about the loot inside, and it only holds while the zones are busy.

Scored honestly that makes a repeat run a fallback rather than a plan. It still pays in the same 3 lines — vendor coin from the bags, materials off the packs the group chooses to kill, and the chance of a saleable piece — and none of those lines grew because the open world filled up. What moved is the comparison, since a contested gathering hour loses the steadiness that is normally its main advantage. Check the zones you actually work before committing the evening, and keep the standing bills of levelling in view, because a fallback that drops nothing saleable leaves the same bills to pay. We do not have figures for how contested the levelling zones are at any given hour, and we are not inventing them.

The part that settles the decision is the hour that never appears in a drop table. Forming a group, waiting on the last slot and travelling to the entrance can take the better part of an hour before the first pull, and that time is spent whether or not anything saleable lands. Set it against the levelling income plan by level, where the same hour produces something sellable every time, and a dungeon only wins when the evening is about slots covered, a chain finished or a recipe learned. We do not have confirmed experience values for either side of that comparison and are not treating the forum figures as official.

Three parts of that hour are worth naming, because none of them appear in the clear itself. There is no group finder to queue into, so the group is put together by hand in chat and the last slot is often what decides whether the evening happens at all. Meeting stones do not summon anybody, so all 5 players travel to the entrance under their own steam and the group waits on whoever is furthest away. And the quest chains that make the trip worth it start outside the instance — the Hall of Thanes chain begins with Beldin Steelgrill at Steelgrill Depot — so a first visit includes a detour to collect the turn-ins before anyone pulls, and a player who skips it clears the place for vendor coin only.

The waiting is not shared evenly, which is why the same run costs different players different amounts of evening. Whoever can tank or heal is taken as soon as they ask; a fifth damage dealer sits in chat while the other 4 slots fill around them, and that time is unpaid in a way a gathering route never is. Price a clear as assembly plus travel plus the run plus the repair at the end, and set that whole total against the standing bills of levelling rather than timing the clear alone. We do not have figures for how long groups are taking to form on any given evening, and we are not inventing them.

On the beta the missing slot is almost always the same one. Trade chat carries a steady run of LF1M TANK messages, and groups that have 4 players ready sit at the stone waiting on the 1 role nobody wants to level in. That shortage is the main thing standing between a formed group and a first pull, so a damage dealer with an evening to spend is not choosing between a clear and a gathering route on equal terms: 1 of the 2 options cannot start until a stranger with a shield agrees to it.

It matters for the arithmetic because income from a dungeon is usually quoted per hour, and the quote assumes the group already exists. Count the wait as part of the hour and the comparison moves, since the levelling income plan by level starts producing the moment you ride out and a clear produces nothing until the fifth player turns up. A tank or a healer loses very little of that hour; a fourth or fifth damage dealer can lose most of it. The practical response is to form the group first and pick the evening's activity second, so the waiting is spent on a gathering route rather than standing in a capital. This is what we have seen in beta chat rather than a measurement, and we do not have figures for how long the tank slot takes to fill.

One response to that shortage is to stop waiting for the missing role altogether. Class-stacked groups turn up in beta chat from time to time, and the warlock-only run is the version people mention most, since every member brings a pet that can hold a pack and the group is complete the moment 5 players agree to it. Those compositions are not quicker inside the instance. Pulls are smaller, the route is taken a pack at a time, and a clear put together that way runs longer than a conventional group would need for the same bosses. What they save is the hour before the first pull, which is the part of the evening the missing role costs.

With experience from the mobs inside reduced and the additional experience on dungeon quests cut by 50%, the longer clear costs nothing that the run was going in for. The payout is the same list it was: slots covered by quest rewards, patterns off the bosses, bind-on-equip pieces a stranger will pay for, and vendor coin from the rest of the bags once the repair bill is settled. A slower route only pushes more of the evening into the clear itself, so the question is whether the bags at the end still cover the standing bills of levelling. We have seen these groups described in beta chat rather than measured, and we do not have figures for how much longer such a clear takes.

That also closes a side trade that used to sit on top of the same activity. Selling runs to a lower-level character was never really a sale of loot; the buyer was paying for the bar at the bottom of the screen, and the seller was paid for how fast it filled. With the experience half cut, there is very little left to sell: the paying character still has to be present for the drops, still rolls against the same table, and comes out with the same gear they would have collected in a normal group. A high-level character looking for a steady income stream has to look elsewhere now, because the product that trade was built on is the one thing a dungeon no longer hands over in quantity.

What remains on the income side of a run is unglamorous and easy to state: bind-on-equip items a stranger will pay for, recipes and blueprints off the bosses, and the vendor coin from everything else in the bags at the end. That is a real total, but it is a flat one, and it does not grow by going back in. Anyone budgeting against a deadline should treat a dungeon as a fixed payout for a fixed clear rather than an activity that scales with hours, and keep the standing bills of levelling in view when deciding whether the evening is better spent on a gathering route instead.

So the reason to go in has moved. You enter for items and for the coin they turn into, not for the bar at the bottom of the screen. That reframes the whole time calculation on this guide: a run is no longer competing with questing on levelling speed, it is competing with a gathering route on what ends up in your bags at the end of the evening. Judge it on slots covered, recipes learned and bind-on-equip items that a stranger will pay for, and the decision gets a lot simpler than arguing about which activity levels faster.

Stated plainly, the instance has changed category rather than value. With experience from the mobs inside trimmed and the additional experience on dungeon quests cut by 50%, a clear is a loot source that sits beside the levelling route instead of being the route itself. Characters still walk in for slots, patterns and saleable pieces; they no longer walk in because it is the quickest way to the next level. Nothing about that makes a run pointless, it just moves it out of the levelling column of the plan and into the income column, where it has to be scored against everything else that fills a bag.

Scored that way, the 2 activities fail in different places. A gathering hour pays a small amount every hour and almost never pays nothing, because ore, herbs and skins come off a route at a steady rate and sell to strangers without a group. A dungeon hour pays nothing at all on an evening where no bind-on-equip piece drops and no recipe lands, and pays several times a gathering hour on the evening one does. The first type of income covers the standing bills of levelling as they arrive; the second covers a large purchase if you can wait for it. If the coin is needed by a fixed date, the levelling income plan by level is the one that behaves predictably, and the run is the thing you do when gear or a pattern is the actual goal. We do not have figures for either experience cut or for what the saleable drops fetch, and we are not inventing them.

With both halves of the experience trimmed — the mobs inside and the quests attached — the question groups ask before forming has changed shape. It used to be how many levels the evening would produce; it is now what the evening leaves in the bags. That is a loot and gold question, and it is answered by the drop table, the quest rewards and whatever a stranger will pay for the bind-on-equip pieces, not by the rate at which the bar moves. A dungeon that covers slots and hands over recipes is worth the trip on those terms alone; one that does neither is an hour you could have spent on a gathering route.

The sharpest form of the complaint in those threads is a question rather than a demand. Once experience stopped being an argument for walking in, players started asking what 45 minutes of 5 people's evening is actually buying, and the answer they give themselves is short: the item that drops and the coin left in the bags after the repair bill. That is a harsher way of putting the same arithmetic already set out above, and it is the version worth using when deciding whether to form a group, because it counts the assembly and the travel as part of the 45 minutes rather than as something outside them.

Scored on those 2 lines a clear either pays or it does not, and there is no third line to fall back on. A chain still to finish puts a blue and a covered slot on the paying side; a repeat clear with nothing saleable in the bags leaves the evening at vendor coin minus repairs, which is the outcome the question is really pointing at. Set that total against the standing bills of levelling before committing, and against the levelling income plan by level if the coin is wanted by a fixed date. The 45 minutes is the figure players use for a clear in those threads rather than a measured one, and we do not have drop rates for the saleable pieces or figures for either experience cut.

The beta reaction is the signal worth watching here, and we are treating it as a signal rather than as data. The complaints we have read are not about the clear being harder, they are about the reason for the clear having moved, and the follow-up question in those threads is always about what drops rather than what it awards. That is a reasonable read of where things stand, but it is player sentiment during testing and nothing has been settled by it. If you are planning an evening against the standing bills of levelling, budget the run as a fixed payout for a fixed clear and leave the levelling calculation out of it entirely.

The one part of a dungeon that still pays experience is the quest log. The chains attached to an instance award experience on turn-in the way any quest does, and that is untouched by the cut to experience from clearing the place. It is another argument for finishing a chain rather than abandoning it halfway: the same turn-ins that hand over the blue at the end also hand over the experience, so an Alliance character working through the 5 quests at Hall of Thanes is being paid in 3 currencies at once while a repeat clear pays in 1. We do not have figures for what those turn-ins award and are not going to invent them. If the evening is meant to produce coin on a deadline rather than gear and levels together, the levelling income plan by level remains the steadier route.

For a levelling character that leaves 2 questions rather than 1. The first is whether the run covers slots you would otherwise buy, which the quest chain answers. The second is whether the hours spent in there beat the hours spent outside, and that comparison is now against questing experience rather than against another dungeon. If the goal for the evening is coin against a deadline, the honest answer is often no, because the standing bills of levelling arrive on their own schedule and a repeat clear no longer moves you closer to the level where they get easier.

There is a second kind of income here that never shows up as coin. Wailing Caverns and Hall of Thanes both hand out bind-on-pickup recipes and blue items alongside the drops you can resell. Those cannot be traded, so they add nothing to the evening's takings, but they still replace a purchase: a blue off a boss is a slot you do not buy, and a recipe learned on the spot is a pattern you do not pay a crafter for. Counting a run purely in gold looted understates it, and counting it as pure profit overstates it too, because the value only exists for the character who was standing there.

The same drops push the other way on the crafting side. Every group that clears those 2 dungeons walks out with soulbound gear covering slots a crafter would otherwise fill, and with recipes spread across more hands than before. That thins demand for the crafted equivalents at exactly the levels where levelling characters would have been buying them, which is worth weighing before committing to a profession that sells to strangers. We do not have drop rates for these items and are not guessing at them; the point is that the goods compete with your order book as well as filling your bags.

The same logic decides what you do with the green bind-on-equip items that fall out of a run. Hall of Thanes, Deadmines and Ruins of Lordaeron all hand out blue gear and blue quest rewards at levels where the auction house is otherwise full of greens, and a buyer who can get a blue for a slot by clearing a dungeon once has no reason to bid on an ordinary green for the same slot. The supply of greens is large, the demand for them sits underneath a free alternative, and both of those facts point the same way.

So the practical handling is dull: vendor the greens, list the blues and the recipes. A green on the auction house costs a deposit, occupies a listing for its full duration and then comes back to you unsold more often than not, while the vendor pays on the spot and the coin is in your bags before the repair bill. The exceptions are the usual ones — a green with a stat block that a twink or a specific class wants, or one that feeds a profession — and those you will recognise because somebody asks for them in chat rather than because they look valuable in your bags. If the evening is meant to produce coin rather than gear, the levelling income plan by level is still the more predictable route, and dungeon greens are a side effect of the run, not a product of it.

One thing undercuts that sorting rule more often than it used to: Forever has rewritten the properties of part of the older drop pool, so an item is not necessarily the item you remember from a decade of reference sites. Martyr's Chain from Blackfathom Deeps is the example people point at — same name, same boss, different stat block. Whatever you knew about which low-level pieces were worth listing and which were vendor fodder was built on the old numbers, and a share of it no longer holds.

The practical handling is to read the tooltip in game before deciding what a piece is, rather than pricing it from memory or from an old database entry. That applies both ways: a drop that was ignored for years can now cover a slot somebody is actually shopping for, and one that had a reputation can turn out to be ordinary. It also affects what a crafter is competing with at those levels, which is worth checking before committing to which profession pays its way. We do not have a list of which items were changed or what the new properties are, and we are not guessing at either.

There is one income line people expect from dungeon drops that does not exist here: appearances. The look of an item is collected by the character the moment it drops, so you keep the appearance whether you vendor the piece, list it or hand it to somebody else in the group. That removes the usual reason a stranger pays over the odds for a plain item, because they cannot buy the appearance from you — they have to see it drop for themselves. A run therefore produces no second market in looks sitting on top of the gear market, and an ugly green with good stats and a handsome green with bad stats are worth what their stats are worth.

Practically that keeps the sorting rule above intact. Do not hold a drop back because it looks unusual, and do not price one as a collector item, since the collector in question already got what they wanted when the boss died. The only things in a dungeon worth listing are still the pieces somebody wants to wear for its numbers, the recipes and the blueprints, and anything left over is vendor fodder against the repair bill.

Questions

Do you make gold running dungeons in Forever?

Some, and most of it is not the coin on the corpses. It is the items that can be sold on: the bind-on-equip drops, the cloth, the recipes. The silver that drops directly barely covers repairs.

What is a BoE and why does it cause arguments?

Bind on equip: an item that can be traded or auctioned until somebody puts it on. It is the most valuable thing a levelling dungeon produces, which is exactly why players argue about who is allowed to roll Need on it.

There is a live proposal on the official forum to bind such items the moment Need is rolled, so they cannot be taken for resale. Nothing has been announced; it is a request, not a change.

Why do I keep running out of gold while levelling?

Because abilities are not free. Every few levels a trainer wants paying, and that bill arrives whether or not you found anything worth selling. It is the single most common reason a levelling character is broke, and it is the reason the BoE argument gets heated — one player's resale is another player's training fee.

Sources