Quest platforms and airdrop marketing: what you're actually buying

A quest campaign looks like advertising if you squint. There is a budget, a target audience, a dashboard that fills up with a satisfying number. But it is not advertising, and pretending it is will cost you a treasury. Quests and airdrops are incentive design: you pay people to do a thing, and they do the thing. That is a real tool with a real job. The trouble starts when the invoice says “growth” and you go looking, six months later, for the users you thought you bought.

What a quest is genuinely good for

Give the tool its due, because it earns it in a specific place. Some products are broken until they are busy. A marketplace with no listings, an order book with no depth, a lending pool with nothing to borrow: the emptiness itself is the obstacle, and no advertising fixes a cold start, because the first honest user arrives, sees a ghost town, and leaves. This is where a quest is the correct instrument. You pay to manufacture the first thousand trades so the thousand-and-first happens on its own. Bootstrapping activity past the point where thinness repels people is a genuine service that ads cannot perform.

Quests do two other things well. They reward a community that already showed up, a fair way to hand value back to the people who believed early. And they create a moment: a campaign, a countdown, a reason for the timeline to talk about you this week rather than next. Platforms like Galxe, Layer3, and Zealy exist because that packaging is legitimately useful. None of this is the problem. The problem is what you ask the tool to do after the moment ends.

The population it selects for

Here is the structural fact, and it is not a moral complaint about farmers. A quest rewards effort with tokens, so it attracts the people best at converting effort into tokens. Those people are professionals. They run many wallets, they read the task list the way a contractor reads a spec, they complete the minimum that clears the reward, and they leave the instant emissions stop. This is not a bug in any single campaign, it is what an efficient market does. The better the questing ecosystem gets at matching effort to reward, the more precisely it selects for mercenaries, because efficiency and mercenariness are the same arrow pointing the same way.

So a quest can buy you activity, reliably, on demand. What it structurally cannot buy you is durable users, because the mechanism filters them out at the door. The person who would have loved your product and stayed for years is sitting in the same funnel as a farmer with forty wallets, and the funnel pays them both the same and teaches you nothing about which was which. You did not acquire a base. You rented a crowd.

The economics nobody prints on the deck

Now the money. You pay for this activity in your own token, at the moment you can least afford to: emissions land on your chart exactly when a wave of freshly rewarded wallets is deciding whether to hold or dump, and most of them were never going to hold. The cost is not just the tokens, it is the sell pressure they manufacture, arriving on schedule. You bought a spike in activity and a dip in price with the same transaction.

Against that cost, one number decides whether any of it was worth it: how many of those wallets are still active after the rewards end. Retention past emissions is the entire ballgame, and it is almost never reported, because it is almost never flattering. The dashboard shows completions, which are guaranteed, and stays quiet about thirty-day survival, which is not. If a campaign will not tell you its post-reward retention, it is telling you its post-reward retention. A completion count is a receipt for money spent, not evidence of a user gained.

The second structural leak: one human, many wallets

Even the activity number is softer than it looks, because of sybil. One person with forty wallets is forty completions on your dashboard and one human in the world, and every one of those wallets collects. Quest platforms know this and run sybil filters against it, and the good ones try hard. But notice the shape of the fight: the filter is swimming against the campaign’s own economics. The reward begs people to make more wallets, and the filter begs them not to, inside the same product. You are paying to attract the exact behavior you are then paying to detect. That is a war you can fight to a draw at best.

Advertising has the cleaner version of this answer because it is not fighting its own incentives. When you bill on outcomes rather than completions, sybil resistance stops being a filter you bolt on and becomes the thing you are buying. Specify ID resolves wallets and devices to one human, so the same person cannot appear as ten conversions and you never pay for the same user twice. A conversion-billed network has no reason to reward extra wallets in the first place, so identity works with the economics instead of against them. This is also why a verified onchain conversion means something a task completion does not: it is one real human doing one action you chose to value.

A decision rule you can actually use

Strip away the argument and you are left with a clean fork. If your goal is activity numbers, and the emptiness of your product is the real obstacle, and you have priced the decay honestly (the tokens, the sell pressure, the wallets that vanish on schedule), then run the quest. It is the right tool and you should not feel bad about it.

If your goal is users who stay, a quest cannot get you there no matter how well it is run, because the mechanism selects against the outcome. For that you want verified conversions from behaviorally-targeted audiences: people chosen because their onchain history across 6,000+ protocols says they are the sort who does the thing and keeps doing it, billed on a CPA against actions that settle onchain, with lifetime volume tracked per converted wallet so you find out months later whether the base you bought is real. That is the number a quest can never show you and this can, because the wallets stay visible and you keep watching.

The play that uses both

These are not rivals, and the best operators run them in sequence. Quest for the cold start, when thinness is the enemy and you need the order book to stop looking abandoned. Then let performance ads build the durable base underneath the activity, aimed at the users who were never going to show up for a task list. And put a holdout on both, because the one question that unmasks everything is whether these users would have come anyway. Incrementality is just keeping an identical audience unexposed and comparing, and crypto wallets convert organically all the time, so without a control group every channel bills you for people you already had. When CoW Swap ran it, exposed users converted at 6x the holdout across more than 1,500 verified conversions and over $50M in attributed volume. A quest cannot pass that test, because it cannot withhold the reward from half its audience and still be a quest. A performance campaign can, which is the whole difference between renting a crowd and knowing what you bought.

The honest position is the same one we take on KOL marketing: keep the tool, give it the job it can actually do, and stop asking it to be the other thing. Quests are a fine way to make an empty room look busy for a while, and a terrible way to fill it with people who stay. To find out what the second kind of spend buys you, a $1,000 test campaign runs the holdout and hands you the retention number either way it breaks.