Every airdrop I have audited was a marketing line item. The airdrop began as a community reward and grew into a full acquisition channel, and the teams that acknowledge it from the start are the ones whose numbers hold up six months later. The ones that describe the drop as a gift to the community are usually the ones asking why their token sits 80% below the listing price by winter.
An airdrop is a paid acquisition channel
Treat the drop as a media buy and the unit economics get honest fast. A campaign distributing tokens to 200,000 wallets at a headline value of $500 each has spent 100 million dollars of treasury on acquisition. Compare that to paid installs or influencer retainers and the conversation changes. Suddenly the marketing team needs a cost per retained user, which is the number that decides everything downstream.
The first cost that shows up is Sybil. When Arbitrum distributed its ARB token to more than 600,000 wallets in March 2023, on-chain researchers documented large clusters of wallets funded from the same source and behaving identically. Every funded farm multiplies your spend without multiplying your audience.
Wallet quality is measurable before the snapshot lands. Funding-source clustering, wallet age, prior protocol usage, and behavior patterns separate first-time users from professional farmers, and the filtering gets cheaper the earlier it runs. Teams that skip it pay retail price for inventory the farmers were always going to flip.
The claim window shows who actually showed up
Watch the first two weeks of trading and the campaign reveals its real audience. Uniswap dropped 400 UNI to roughly 250,000 wallets in September 2020, worth about $1,200 per wallet at the time. Dashboards tracking those wallets on Dune and Nansen showed exchange deposits climbing within days of the claim opening, and a large share of the claimed supply reached market makers before the month ended. That is normal behavior. People who were farming for cash will sell for cash.
The useful metric here is claim-to-deposit velocity. Fast velocity means you bought traders. Slow velocity with actual app usage behind it means you bought users, and those are priced very differently at the next board meeting.
Retention gets budgeted later, if at all
The ninety-day number is the one that decides whether the campaign worked. A claim spike is a launch event. What survives is whether the wallets come back, vote, trade, or integrate the product after the sell pressure clears. Projects that plan this part use task-gated claims through platforms like Galxe or Layer3, vesting schedules that extend past the initial unlock, and in-app rewards that only pay out to returning users.
The teams that skip it end up buying the same users twice: once with the drop, and again with a second campaign aimed at the audience the first one failed to keep.
Three questions before the budget gets signed
Ask these before the tokenomics deck gets approved. What does one user who is still active on day 90 cost? What share of claimers is estimated to be Sybil, and what is the plan if that share doubles? And what happens to price when the largest unlock lands?
An airdrop can buy a launch week, and done well it can buy a real community. The difference is a retention budget, a Sybil filter, and the honesty to call the whole thing what it is: paid acquisition, wearing a whitepaper.