Lab: free tools · size first · stock scanner · crypto scanner · journal
Protocol Retention Metrics
Protocol Retention Metrics is a crypto market concept used to structure analysis, execution, and risk decisions with measurable rules. It helps practitioners translate noisy data into consistent portfolio actions over time.
Retention metrics show whether users keep coming back to a protocol after their first use. Cohort charts track groups of new users over weeks or months. Strong retention suggests real product demand; weak retention after an airdrop or incentive program suggests rented activity. Retention is harder to fake than raw user counts, which makes it a useful check on headline growth.
How a cohort table works
Group wallets by the week they first used the protocol. Then, for each later week, count what share of that group came back and did something. Say 1,000 wallets first swapped in a given week. If 300 return in week two, 180 in week four and 150 in week twelve, the curve is flattening, which is what real users look like. If it drops to 20 by week four and stays near zero, those wallets came for something other than the product. The same table can be built with volume or fees in place of wallet counts, which shows whether the people who stay are the ones who matter.
What muddies the numbers
A wallet isn't a person. One farmer can run hundreds of wallets to collect an airdrop, and one person may use three wallets for privacy. Points programs reward repeat activity, so retention can look strong right up until the snapshot, then fall off a cliff. Blur's early volume is a good example of activity that came largely from wash trading and point chasing. Bots, contract wallets and aggregators routing through a protocol can all count as users too.
Using retention in research
Look at cohorts from before, during and after any incentive program and compare them. A protocol whose cohorts hold up after rewards end has something people want. Check retention by fees paid as well as wallet counts. If a team only shows total users or total volume and won't share cohort data, that tells you something too. Free dashboards built on on-chain data let anyone check these numbers without relying on the team's slides.