Test paywall design and plan mix first; leave price tests for later

Jonathan ParraYeah, multi-page paywalls uh sometimes work well. I like to test them, and uh b- and I like them because you can add different things to the different pages. So maybe instead of bullets here, I could have tried a table instead of bullet list. Um I could have moved these accolades to this next screen. Um you know, there there's so much that you can do um in your design test.

Jonathan ParraSo design tests really are the bread and butter of anyone that wants to improve their KPIs with A/B testing for uh paywalls. You know, you'll get more mileage out of design tests and price packaging testing than really anything else.

Jonathan Parra… some apps that they're um they just started A/B testing, and then they have all of these uh uh pricing that they added, um and then they immediately want to start experimenting with pricing. Uh but what that does is it actually creates quite a bit pain in the ass to manage, because uh in Superwall at least, you have to have all those users um in an audience in a specific cohort, so you have to put them uh use user seeds to kind of manage them.

Jonathan ParraAnd then you have to manage them uh for every time that you want to run an experiment. And so it it just becomes a little bit more difficult to manage once you start doing price testing. So it's not something that I like to do immediately, it's it's something that I like to do kind of later down the line after we've ran a lot of uh of tests.

Joseph ChoiThat's really helpful. I mean, that's crazy. I mean, it's interesting that you have like the insight or the the most surprising insight was that there is no one paywall, um but there still sort of is a one starter paywall that you can kind of use as, you know, it's not going to be, you know, the absolute best like out of any paywall in the world, but it will be like it sounds like around like 80% there, and then you can like optimize it further.