I'd like to track the number of new paying customers we get each month (to calculate CAC). I’ve found two ways to do this, but am seeing large discrepancies between the two. Can someone help me understand the discrepancy, and recommend the best way to get this number?
Here is what I’ve been looking at:
- Active Subscriptions Movement chart - # of “New Actives” for a given month.
- Customers list - filtered down (e.g. for May, first purchase date >= 2022-05-01, first purchase date < 2022-06-01, total spent > $0)
I’m seeing a discrepancy of 13% for May, and 50% for June (numbers are in the low hundreds).
My first guess is data lag. This comment suggests that the Customer list can take 2 hours to refresh, but that wouldn’t explain the magnitude of the difference.
is there something I’m missing in terms of the definitions for “active subscriptions” and “first purchase date”?
From the “Active Subscriptions” chart info on hover:
Active Subscriptions shows the number of unexpired, paid subscriptions at the end of each period charted.
I’m assuming this would not include trials.
Appreciate any help/clarification I can get here - thanks!
Best answer by ryanView original
I might suggest using the Conversion to Paying chart for this calculation: https://demo.revenuecat.com/charts/conversion_to_paying?chart_type=Line&customer_lifetime=30%20days&range=Last%2090%20days%3A2022-04-01%3A2022-06-29
The “Paying Customers” metric in that chart will be cohorted by the customer install date instead of the purchase date. This is usually preferable for a CAC calculation since you want to know how many customers you acquired in a timeframe eventually became paying. By looking at the Active Subscriptions Movement chart you’re looking at purchase date, but you don’t know for sure when you acquired those customers that ended up purchasing.
I think the reason your Customer List doesn’t match up is because the total_spend filter is not timeboxed to the date range you set for first purchase date, so you’re building a list of customers that started a trial in that timeframe and became paying later. You should be able to click into a few profiles to spot check.
A customer list that might be useful is replacing the purchase_date filter with first_seen, and keeping the total_spend > $0. This would show you customers that installed within a date range and the revenue generated by that cohort.
This is super helpful, thanks