# External Payment Link vs In-App Purchase: RevenueCat’s $40k Experiment Explained

Channel: Sub Club by RevenueCat
Video: https://www.youtube.com/watch?v=R3gbKaowcq4
Duration: 13 min
Language: English
Words: 2196
Transcript page: https://viewrankai.com/tools/youtube-transcript/R3gbKaowcq4

---

[0:00] Hello everyone. I've got some really interesting data to share with you about an experiment we recently did at RevenueCat using the web payw wall links, the external links that are now available on the app store. For those of you who maybe haven't been paying attention or don't follow the app store and the app economy very closely, a

[0:21] judge recently updated an injunction. And in it, the judge opened up the door for developers to use external payment links from within an app. That means inside the US, you can link out to an external payment source. You don't have to show what the industry is calling a scare sheet. Apple, you know, had a sheet that would pop up when you'd go to an external link that would say, you know, this is not Apple Payments and, you know, we're not liable for fraud. At RevenueCat, we acquired an app called Dipsy. We acquired this app specifically to run these kind of experiments to beta test our revenue cap features. Within

[1:01] days of the ruling, my colleagues built out a system to take advantage of this a button that would link to external purchases. And of course, we already have a payw walls product. We already have an experiments product. We already have targeting and placements products that allowed this to happen. So, first off, let's talk through the different variants we used. So, we have A, B, C, and D. A fourvariant test. Now, as I go through these, you're probably going to be yelling back at the screen. You know, why didn't you do this? You know, why is the wording that? You clearly skewed the results. So, let me know in the comments

[1:35] what you think we should change, what you think could improve the performance of each of these different payw walls. But what we wanted to do was get a baseline. We didn't want to skew it in any one direction or the other. We just wanted to take a payw wall that Dipsy had already tested and knew performed fairly well. This was kind of their winning payw wall after lots of paywall iteration. They were still using a React Native paywall in the app. They had not

[1:59] migrated yet to our payw wall product. So the first variant is their React Native paywall. The second variant, variant B, is actually a recreation of that payw wall in the Revenue Cat payw wall product. So we did want to make sure that we could compare directly the performance of our paywall product against the existing payw wall so that using our paywall product didn't skew the results. Now you will notice there are some differences. So in the original payw wall the font size was different because our paywall product didn't perfectly represent the font size that they originally used in React Native. So there are subtle differences but if you

[2:36] look at these two they're very similar. Variant C is where we wanted to just give users choice. So this variant is every option. You can select annual or monthly and then hit the blue button to check out with inapp purchase or you can tap the red button to try for free and save 30%. Now that takes you out to the web. We've already been discussing internally that you know maybe we could have made this more clear. Save 30% on the web. There's a lot of different ways we could iterate on this payw wall to maybe, you know, skewed the results toward, you know, making it better. But

[3:13] but we thought this was a good baseline using the exact same wording. And so the wording is exactly the same except for the and save 30%. Variant D actually only sends people to the web. So again, we tried to recreate the payw wall as close as possible. It's exactly the B variant payw wall. Only difference in this payw wall is that when you tap the try for free button, it actually takes you out to the web. So that's kind of an overview of the different variants. So let's get into the results. And just to explain this chart a little bit, the inapp purchase only would be variants A

[3:48] and B. And then the inapp purchase plus web would be variant C. So this red is A and B, green is C and yellow is D. So the web only this is proceeds per user after deducting fees. So we estimate Dipsy's all-in cost on the web to be about 6%. So the revenue per user on the

[4:17] web only payw wall variant D was $1.96. on the inapp purchase only that was $29 again taking into account Apple's 30% fee. So Dipsy are currently paying 30%. So this is kind of the realistic four dipsy results of the experiment is that it it's pretty close to a wash. It's about a 6% drop in proceeds on the web versus the inapp purchase. And then the inapp purchase plus web overall did underperform just a tad. when you go to the 15% fee from Apple. So this would be, you know, if you are still in the small business program only paying Apple 15%. It's going to be pretty tough to

[5:02] make the numbers work on external purchases. You know, this is a massive drop in revenue per user uh sending them to the web versus only using AP. So, let's dig a little deeper into those results and see why it ended up being about a wash on that 30%. So, this is Revenue Cat's experiments product. This is what we used to run the test obviously. And what you'll notice right off the bat in this chart is that we did see a huge drop off in trials started in variant D. So, the web only people who linked out to the web were just far less likely to start a trial. And this is

[5:42] pretty consistent with what we've seen and heard around the industry is that even with the ease of Apple Pay, even with, you know, people having their credit cards stored now with Apple, people just don't start free trials at the same rate on the web as they do in app. And so we saw this huge dip in trials started. But then of course and anybody you know in this industry will see that the trial conversion rate is much higher on the web than it is in the app. So that made up for the lower trial start rate. More trials did convert even though fewer trials were started. And so

[6:24] when you look at conversion to paying the total number of paid users, you do still see that increase for variant B that variant was the best variant overall just sending people right to inapp purchase. But that's not accounting for the difference in the fee. And so that's where things really washed out was that at a 30% fee, even though more com people converted to paying, we did see that accounting for the fees, it ended up being closer to a wash. And so one of the fascinating things about this experiment is in variant C where we gave customers the choice of going to the web with a 30%

[7:06] discount only 68 out of 203 actually started a trial and then converted from that free trial. Only 68 compared to 75 selected the annual plan inside the app and then 57 selected the monthly plan inside the app. So if we go back and look at the variance and again let me know in the comments how you think we could improve this but this was again designed to be the ultimate kind of customer choice. You would select between annual and monthly hit the blue button to start or save 30% by going to the web. And so even with that incentive of 30% far more people more than twice

[7:46] as many people started a free trial and ultimately converted to paying on the web compared to in the app. So, I've been trying to figure out what this is. You know, without doing qualitative research, asking people why, you know, maybe there's a preference for inapp purchase. Maybe it's just the path of leash resistance. You know, maybe we should have added that monthly option because we do see a lot of people did choose the monthly option that is a lower price to get started. There's a lot of things we could test next to see, you know, how we could improve this, but it is just striking that. So few people

[8:24] went to the web even with that 30% discount. Now if you see on our web only payw wall and if we go back to that payw wall you can hit the view all plans buttons just like you could any of the other variants except for variant C. And on the view all plans you could select the monthly or the quarterly. And so you do see in the conversion to paying on the web that you know the majority of people did stick with that annual plan but some did switch to the monthly plan and some did switch to the quarterly plan. So really fascinating results here. Now if we switch the chart to

[9:02] realized LTV. So this is the amount of revenue each variant has generated in total. You can see right off the bat, the math is actually pretty easy right now is that we have $10,000 of revenue made on variant B and then only 6800 on variant D. So it's almost exactly that 30% app store fee. So again, it just shows the wash of sending people out to the web. All right, there's one more bit of data I wanted to look at to better understand these results, and that is subscription status. Cuz I know a lot of you were thinking when I was showing the results previously that, oh yeah, it's a

[9:42] wash today, but web subscriptions renew at a higher rate. And spoiler alert, you're probably right. But let's look at the data. So this chart shows the current subscription status. So for an active subscription that's going to renew at some point in the future, what's the current status of autorenew? Are they set to renew or are they set to cancel? So, if we look at May 2026, what we're looking at is any annual subscription that was started in May 2025 that could potentially renew in May 2026. And of those that could potentially renew in 2026, how many currently have autorenew turned on and how many currently have autorenew turned

[10:31] off? So for all subscriptions in May who were build on the web, only three and a half% are currently set to cancel, only three and a half have gone into settings and turned off autorenew on the web. All right, so now let's look at the app store data. And as you might expect, if you work in the industry and have ever experimented with web billing, you'll see you people on the app store do turn off autorenew at much higher rates. So, of that cohort that that could potentially renew in 2026, almost 19% have already turned off autorenew. So, this is an early signal. You know, we

[11:12] won't know exactly how many end up renewing until May of next year. And actually, I'd love to do a follow-up post then to get exact numbers on how things turned out. But this is a signal that even though it was a wash or you know slightly favorable to inapp purchase on the initial conversion the LTV is likely to be higher on those web purchases because 19% already having turned off autorenew compared to 3.5% on the web billing having turned off autorenew that is skewing toward the favor of web billing. All right, so

[11:49] those are the results as of May 2025. Things may continue to change. We will be launching more experiments. As I said, please leave comments below as to how we could improve our conversion on the web, any other suggestions on paywall improvements, anything else we should be experimenting with and sharing the data. Again, we acquired Dipsy specifically to be able to do these kind of experiments and share the data with our customers and with the broader community. And that's ultimately how you're going to maximize the revenue from your subscription app is to take a hybrid approach. Don't send everybody to the web. Don't leave everybody in the

[12:27] app. Well, if you're in the small business program for now, you probably do still want to keep everybody in the app using inapp purchase. But if you're you're not in the small business program and you are paying 30%. I think the ultimate solution is going to be this hybrid approach where you send some folks to the web and keep some people in

[12:44] the app. Thanks so much for joining me. Cool. [Music] Hey, jump. [Music]

---

About this transcript
Read from YouTube's own caption track and laid out by ViewRank AI (https://viewrankai.com).
ViewRank AI finds the videos already beating a creator's own average on Instagram, TikTok and YouTube Shorts, transcribes them from the audio itself in more than 60 languages, and turns what worked into new ideas and scripts.
Free transcript tools, no account needed: https://viewrankai.com/tools
How to read any video this way: https://viewrankai.com/llms.txt