Understanding Retention in Crypto
The thing that matters the most but talked about the least
If there’s one metric that investors should be looking at much more carefully in crypto is retention, but more specifically the nuances of the data points that make up the quality of retention. I’ll explain what I mean below.
Let’s take a simple dApp for example that says they have a D60 90% retention rate with their users. You’d think that’s great, right?
But now what if I told you that that D60 90% retention rate rate was compromised of users who:
All joined in one cohort
Started with brand new wallets
Have a net-worth of less than $100
Are you really going to look at that D60 90% retention number that carefully? Probably not, because you know they’re getting airdrop farmed. Nice.
Furthermore, what if we knew through the data that out of the 1000 wallets you have, 3/4 of them were creating multiple wallets and then using your product? Your retention numbers are once again thrown off since high quality users who stopped using the product will skew your retention numbers significantly again.
Why does this even matter, if number go up then token go up and yield go up? Right?
False. The problem is that whatever acquisition costs you’ve incurred (from a time, money or resources) to run that campaign became a lot more expensive than your projections or calculations are saying. Here’s a more tangible example to make it more clear:
You spend $10,000 on a campaign to acquire 100 users. You’ve effectively paid $100 per user as your cost to acquire them.
With your $10m in venture funding, you think to yourself “Hm, surely that’s not too bad. We just need to spend $1m to acquire 10,000 users”.
After 90 days, only 5 users remain from your campaign.
Your real cost to acquire a user has gone up 20x, from $100 to $2,000. You realise you’ve now spent $1m to acquire 500 users.
That’s going to go well in your next board meeting, ouch.
Of course some of these numbers are exaggerated, but from the data I’ve seen — not by too much.
The bull market and ponzinomics lured people into a false sense of their economics. The bull market is revealing it but people still don’t know the real extent of it all — when they should.
Before you focus on acquisition too much, you want to make sure you don’t have a super leaky funnel because it’s going to cost you a lot of tokens.
The only reason I’m writing about this is because the more I see through the data with ARCx, the more I realise commonalities in problems across the board.
Please, think about retention and start measuring it properly. Your unit economics will thank you for it.
Kerman Kohli is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.