Judge a web3 marketing strategy by what is still working the month after paid spend ends: retained users, channels you own and partners who keep sending people. Paid crypto distribution can fail that test badly. One academic study of nine major airdrops found that in some, up to 66% of the tokens were sold soon after they were claimed.
Who decides this
The CMO owns the strategy and is accountable for one outcome: a user base that keeps growing at a lower cost per retained user each quarter. The CFO approves the budget and should see it split between spend that rents attention and spend that builds assets. The CEO decides product priorities, because the channels that compound best usually need engineering time, not media budget.
What should the marketing budget buy?
The budget should buy retained users in a named cohort and the assets that keep producing them, not attention. Attention decays as fast as spend stops. Assets keep working.
The pool is smaller than it looks. Andreessen Horowitz's 2025 State of Crypto report estimated monthly active crypto users at 40 million to 70 million, as reported by Decrypt. Every project is bidding for the same people, often through the same creators and quest platforms.
Renting those users again each quarter is the expensive path. A crypto distribution strategy that compounds converts some of them into an audience the project owns and can reach without paying a platform or a creator each time.
What we would do: split the plan into two budgets. One is acquisition, measured on cost per retained user. The other is assets, measured on how many users they deliver without new spend six months later.
The ratio between the two should shift over time. Early on, acquisition will dominate, because there is nothing yet to compound. By the end of the first year, the asset budget should be growing and the acquisition budget shrinking. If the ratio has not moved, the CMO should explain why to the CFO.
Which channels keep working after the spend stops?
The channels that compound are the ones you own or that give a partner a lasting reason to send you users. Product-led distribution, integrations, search content and a community with real work to do all qualify. Paid creators, quests and airdrops generally stop when the money does.
| Channel | Keeps working after spend stops? | What it builds | Main risk |
|---|---|---|---|
| Product-led (referrals inside the product, shareable outputs) | Yes, if the product is used | Owned acquisition loop | Needs engineering time; slow to start |
| Integrations with wallets, chains and protocols | Yes, while the integration is live | A partner who sends users | Partner priorities change |
| Search and reference content | Yes, for years if maintained | Inbound demand you own | Slow; needs subject expertise |
| Community with a defined job (support, testing, governance) | Partly; needs a small paid team | Retention and advocacy | Turns into a price chat if left without a job |
| Paid KOLs and creators | No | Short-term reach | Disclosure rules; audiences rented, not owned |
| Quests and points campaigns | Rarely | Activity counts | Rewards farmers over users |
| Airdrops | Only if tied to continued use | Initial holder base | Fast selling by recipients |
This table is our editorial assessment, not a market dataset. No published study we found ranks crypto channels by durability. The order still holds as a working rule: build at least two rows from the top half before scaling the bottom half.
Integrations deserve more attention than they usually get. A wallet, chain or protocol that surfaces your product to its own users is a channel that keeps running after your budget stops. The price is engineering time and a commercial reason for the partner to keep the integration live. Write that reason down. If it depends on a one-off grant or a co-marketing fee, it will end when the payment does.
How do we design incentives that do not just pay farmers?
Tie each reward to the behaviour you want repeated, pay it over time, and accept a smaller headline number at launch. That trade-off is the core of incentive design.
The evidence on untargeted rewards is poor. The academic study of nine airdrops attributes much of the fast selling to farmers who optimise for eligibility rules. Its case study of one large Layer-2 airdrop found a short spike in activity that did not become sustained use.
A 2024 study of 62 airdrops by the trading firm Keyrock, reported by BeInCrypto, found 88.7% of tokens priced lower after 90 days. The same study suggested larger allocations held up better over longer periods. Generosity was not the problem. Untargeted generosity was.
The trade-offs are real. Vesting and activity conditions cut claim rates and draw complaints. Sybil filters exclude some genuine users. Fewer, better-targeted recipients produce a smaller launch-week chart. We would accept all three in exchange for a holder base that is still active at day 90.
In practice, that means a few specific choices. Reward actions that only a real user takes, such as repeated use over weeks, not one-off transactions. Release rewards in stages tied to continued activity. Cap what any single wallet cluster can earn. Publish the rules before the campaign starts, so the community can see the logic and the CMO can defend it.
How do we know distribution is compounding?
Distribution is compounding when the share of new active users who arrive without paid spend rises month after month, and each cohort's retention curve flattens instead of falling to zero. If neither is true, the budget is renting users.
Track four numbers monthly, from your own systems and on-chain data:
Cohort retention. Of the users acquired in a given month, how many are active at 30, 90 and 180 days. Read it by channel.
Unpaid share. The proportion of new active users who came through product referrals, integrations, search or community rather than paid channels.
Cost per retained user. Spend by channel divided by users still active at 90 days. This is the number the CFO should see.
Partner-sourced users. How many active users each integration delivered. A partner that sends nobody after three months is a press release, not a channel.
Exchanges look at similar signals. Coinbase's listings page says its reviews weigh demand indicators, including token holder traction, and also look at how tokens are distributed. Durable distribution helps the listing case too.
What has to exist before the budget ends?
Before paid spend winds down, the project should own a reachable audience, have at least two integrations sending users, hold a content base that ranks, and run its community without an agency. Plan the hand-back from day one, not the last month.
Community is an asset easily left with an agency. Bring it in-house early with a small team that answers support questions, runs testing programmes and explains governance. Give the community a job beyond price talk, and it becomes a source of referrals and feedback. Leave it without one, and it fills with speculation the project cannot control.
Each asset needs an owner and a date. Write them into the plan beside the paid campaigns they will replace. If an agency runs any of them, its contract should require handing over accounts, data and creator contacts on exit; our guide to crypto marketing agencies covers those terms.
The test is blunt. Turn off paid spend for a month on paper. If new active users fall to near zero in the model, the strategy has not built anything yet, and the next tranche of budget should go to the top half of the table.
What goes wrong: distribution left to chance
The common failure is a strategy that is really a media plan. The project has a budget, an agency and a calendar of campaigns. It has no decision about which users it needs, which channels it will own, or what remains when spend stops.
For two quarters the dashboards look healthy. Followers grow, quests fill and creators post. Then the budget tightens, the agency scales back, and new users fall to almost nothing within weeks.
Nobody had chosen the users, so nobody built the loops that would keep bringing them. The fix is upstream. Define the cohort, fund the assets alongside the campaigns, and measure the unpaid share from the first month.
How Alcyone approaches it
Alcyone builds the distribution plan before choosing channels, and treats paid campaigns as the bridge to owned ones. Its marketing advisory covers the marketing strategy and channel plan and KOL and community activation, and the consumer distribution results in Alcyone's client work show the approach at scale. In one published case, a DePIN consumer app reached about half a million monthly active wallets. Engagements begin with a thirty-minute call with a senior partner and a fixed-scope, fixed-fee assessment.