Melissa Kwan

Our conversion rate dropped from 40% to 12% over the last 3 weeks 😳

We hadn't made any major changes so that sharp of a drop didn't make sense, but it was worrying. We were scratching our heads and couldn't figure it out.

When I decided to manually counted the number of trial to paid conversions, I realized that number was much higher than what our Stripe data showed.

Last month, in efforts to try and prevent people from signing up with credit cards that bounce, we charged cards $1 at sign up and refunded it right away. Turns out when we did that, Stripe counted them as a customer instead of a trial, and therefore no longer started counting those sign ups as "conversions".

Luckily, this was a false alarm. But goes to show how a small test can throw things off.

PS. This experiment didn't end up making a huge difference in delinquents.