We Studied 2,995 Paywalls. Here’s What Actually Converts. — Transcript

Full transcript

1. 0:00After studying nearly 3,000 paywalls,
2. 0:03hundreds of subscription flows on Moby,
3. 0:05and talking to someone who's designed
4. 0:07over 4,700 paywalls,
5. 0:10we found this.
6. 0:11>> [music]
7. 0:12>> Sometimes, an ugly paywall could
8. 0:14outperform a beautiful one.
9. 0:16>> No matter what I did, the ugly paywall
10. 0:18with a ton of text completely over
11. 0:21performed anything that I would do.
12. 0:22>> More friction could increase
13. 0:24conversions.
14. 0:25>> Multi-page paywalls almost always do
15. 0:28better than single-page paywalls.
16. 0:30>> Every paywall today kind of looks the
17. 0:32same. So, I wanted to know what
18. 0:34separates a high-performing paywall from
19. 0:36an average one.
20. 0:38I used to think the decision happens on
21. 0:41a paywall, but actually, users sometimes
22. 0:43decide to pay before they even see the
23. 0:46paywall.
24. 0:46>> All the different touchpoints are
25. 0:48extremely important.
26. 0:49>> The paywall can appear after onboarding,
27. 0:52when you unlock a premium feature,
28. 0:54inside settings, [music]
29. 0:56or even as a win-back offer before you
30. 0:58leave.
31. 0:59>> Those paywalls speak more to the user
32. 1:01because they're more personalized.
33. 1:03You're constantly trying to get the user
34. 1:05to either pay or stay as a subscribing
35. 1:08member, and so you have to think about
36. 1:10the entire life cycle of the customer.
37. 1:12And yes, product is important for being
38. 1:14able to solve a problem, but life cycle
39. 1:16monetization and the paywalls, that's
40. 1:19the only part of the app that makes
41. 1:20money.
42. 1:21>> A paywall isn't just a screen. It's a
43. 1:24flow.
44. 1:25Before asking for money, Opal spends
45. 1:27time selling the outcome, getting 8
46. 1:29years of your life back. By the time you
47. 1:32reach the paywall, it doesn't feel like
48. 1:34[music] an interruption.
49. 1:36By showing users how many years of their
50. 1:37life they could save, trial sign-ups
51. 1:40went up from 7% to 17%.
52. 1:43Jonathan prefers multi-page paywalls.
53. 1:46>> Multi-page paywalls almost always do
54. 1:49better than single-page paywalls. And
55. 1:51so, in this paywall, you're at the end
56. 1:53of the customer journey and onboarding,
57. 1:55and you've just showed the customer that
58. 1:58hey, your plan is ready. This is what
59. 2:01you can expect. You're speaking to them
60. 2:02from an emotional perspective. You're
61. 2:04telling them they're going to start
62. 2:05feeling like themselves in 4 weeks or
63. 2:07you're going to tell them something that
64. 2:08is in line with their actual ideal
65. 2:10outcome. Does the paywall already feel
66. 2:13part of the experience? Like if it's an
67. 2:15onboarding, is it a natural segue to
68. 2:17kind of finish the onboarding?
69. 2:19>> A Head takes a similar approach.
70. 2:22Information unfolds gradually. It gives
71. 2:24users time to process the decision
72. 2:26instead of forcing them to absorb
73. 2:28everything at once.
74. 2:30I think when a paywall works, it doesn't
75. 2:32feel like you've hit a wall. It feels
76. 2:35like the product is asking you to take
77. 2:37the natural next step.
78. 2:39Even if you have a well-designed
79. 2:40paywall, [music] if it appears at the
80. 2:42wrong moment, it's still a bad paywall.
81. 2:45One example is the one-time offer. If
82. 2:48you close it, it's gone. Paired with
83. 2:50animation and haptic feedback, this
84. 2:52helps to pull the attention of the user
85. 2:54the moment that you're considering an
86. 2:56upgrade. Even when the timing is right,
87. 2:58people can still hesitate. Will I forget
88. 3:01to cancel? Am I making the right
89. 3:03decision? And that's why some of the
90. 3:05best-performing paywalls focus on
91. 3:08reducing risk. Blinkist had a problem.
92. 3:10Their users complained about free trials
93. 3:12because they felt tricked into being
94. 3:14charged. So, they redesigned the paywall
95. 3:17screen to show a step-by-step timeline.
96. 3:20The result was more trial sign-ups,
97. 3:22fewer complaints. Push notification
98. 3:25opt-ins also went up, which makes sense
99. 3:28because they remind you before your
100. 3:30trial ends.
101. 3:31>> [music]
102. 3:32>> Tipstop emphasized the free trial, added
103. 3:34a discount badge, and made the offer
104. 3:36easier to understand. The offer didn't
105. 3:38change. The way it was presented [music]
106. 3:41did. Direct conversions almost tripled.
107. 3:44And then there's Slopes. Instead of
108. 3:46redesigning the paywall, they asked a
109. 3:48different question. What if we got rid
110. 3:50of the paywall altogether? They took
111. 3:52inspiration from Apple's redeem free
112. 3:54trial approach and built what they call
113. 3:57a pay ramp. Users see a single action,
114. 4:00redeem your free week. No traditional
115. 4:02paywall standing in the way. This alone
116. 4:05increased trial starts by 25%.
117. 4:09Which brings us to one of Jonathan's
118. 4:11favorite paywall tweaks.
119. 4:12>> This no commitment cancel anytime
120. 4:14subtitle always seems to do well. Adding
121. 4:18it to the paywall will bump things up
122. 4:20incrementally. And then also having some
123. 4:22kind of call to action for what the user
124. 4:26is doing instead of just continue is hit
125. 4:29or miss, but I do like to experiment
126. 4:31with that as well. Another thing too is
127. 4:33like having this right chevron here on
128. 4:36the button. I can't make a judgement yet
129. 4:39as to whether like this in itself does
130. 4:41the best because I have not done a test
131. 4:44yet where everything is the same except
132. 4:45for this right chevron, but in most of
133. 4:48the paywalls that end up winning, they
134. 4:50have this right chevron. So, I've also
135. 4:52started to include it on the button
136. 4:54along with this subtitle here.
137. 4:56>> Jonathan takes this idea one step
138. 4:58further.
139. 4:58>> So, if they try to hit cancel over here,
140. 5:01I like to show the users an exit intent
141. 5:04sheet. If they're not ready to commit
142. 5:05for a year, you can still subscribe to
143. 5:08the monthly plan.
144. 5:09>> And when we talked about aggressive last
145. 5:11minute discounts?
146. 5:13>> I wouldn't really do this unless it's
147. 5:14maybe Cyber Monday Black Friday. I'm
148. 5:17really curious to see how Apple is going
149. 5:19to approach that this year for all these
150. 5:21different sales, but to make sure that
151. 5:23I'm compliant and safe, I'd recommend
152. 5:25instead to have an offer with a larger
153. 5:28trial.
154. 5:28>> Giving users a longer trial could help
155. 5:31you achieve the same goal while
156. 5:32preserving trust. So, if reducing risk
157. 5:35helps with conversions, we would expect
158. 5:38that removing friction will do the same.
159. 5:40But that's not the case.
160. 5:43To start your free trial on Outsider,
161. 5:45>> [music]
162. 5:46>> you'll need to enter your credit card
163. 5:47details. Sign-ups dropped by more than
164. 5:50half, but the conversion rates increased
165. 5:53by five times, more than double the
166. 5:55paying customers.
167. 5:56The extra friction filtered out people
168. 5:59who were never serious in the first
169. 6:00place. Moonly discovered something
170. 6:03similar. They introduced a free trial,
171. 6:05but only on the annual plan.
172. 6:09Sometimes what changes isn't the offer,
173. 6:11it's how the offer is framed. One of the
174. 6:14most common techniques is social proof
175. 6:17with real user reviews and a five-star
176. 6:20rating.
177. 6:24>> These social accolades are really nice
178. 6:26to kind of show credibility and
179. 6:27authority.
180. 6:28>> The second common technique is value
181. 6:30framing.
182. 6:31>> [music]
183. 6:31>> You're framing the value in such a way
184. 6:32that it matches what the users care
185. 6:34about. All these screens are selling
186. 6:37different futures. [music]
187. 6:40And then there's price anchoring.
188. 6:43Tide breaks down the price into smaller
189. 6:45weekly amounts. Ahead compares its
190. 6:48subscription to things people already
191. 6:50spend money on, like coffee or therapy.
192. 6:54Next up, pricing and packaging.
193. 6:56Headspace [music] tested 7-day, 14-day,
194. 7:00and 30-day free trials. The winning
195. 7:02variant was this, 14-day trial for
196. 7:05[music] the annual plan. Even though the
197. 7:07annual plan costs more, the longer trial
198. 7:09made the whole decision feel less risky.
199. 7:12If you're wondering what the best
200. 7:13pricing structure is, [music] it depends
201. 7:16on what you're optimizing for.
202. 7:17>> Ideally, you want to default to a yearly
203. 7:20product because we see that that has the
204. 7:22highest LTV, but a yearly product may
205. 7:24not be the best for your business. If
206. 7:27you need to, you have that yearly
207. 7:28product with [music] a weekly or with a
208. 7:30monthly product. You only want to show
209. 7:33two to decrease cognitive load on the
210. 7:35user. So, if you do want to support
211. 7:37other pricing options, you always hide
212. 7:39it behind a view all plans button that
213. 7:41opens up a sheet with all the different
214. 7:42pricing. You want to keep the base
215. 7:44paywall simple, get a bunch of users in
216. 7:46there to see how your base stats perform
217. 7:49before you go crazy on optimizing all
218. 7:52your different placements.
219. 7:53>> Jonathan is also a big believer in
220. 7:55table-based [music]
221. 7:56paywalls.
222. 7:58>> I am a huge fan of tables. They're a
223. 7:59little hard to build in Superwall. I
224. 8:02think that's maybe why, like in
225. 8:03RevenueCat and Superwall, I'm not really
226. 8:05seeing a lot of tables. These also do a
227. 8:07good job of showcasing what you're
228. 8:08missing out on if you don't subscribe.
229. 8:10This also works pretty well. A video
230. 8:13paywall showing you the app in action
231. 8:16and simple bullet list, simple social
232. 8:18proof here of people already using your
233. 8:20app.
234. 8:20>> Video could communicate value faster
235. 8:22than static screenshots or bullet
236. 8:24points. Now, one thing we kept coming
237. 8:26back to was polish.
238. 8:28>> I think animations work really well.
239. 8:30There's an app that I really like right
240. 8:31now that's going viral. It's called
241. 8:33Focus Town, and they've done an amazing
242. 8:35job of branding and creating a Animal
243. 8:38Crossing-esque experience, and that's
244. 8:40the level of polish that we're going to
245. 8:42be seeing from the most successful apps
246. 8:44moving forward. You know, the bar has
247. 8:46just been raised because now that anyone
248. 8:47can ship an app, the ones that really
249. 8:50put in the time to polish it with
250. 8:52animations, with custom character
251. 8:53designs, with world building are going
252. 8:55to be the most successful moving
253. 8:57forward.
254. 8:58>> When two products are competing for the
255. 9:00same user, polish could be the reason
256. 9:02why someone chooses your product over
257. 9:05the others. [music] If you're building
258. 9:06with AI, that's exactly what Mobbin MCP
259. 9:09is for. It gives tools like Claude,
260. 9:12ChatGPT, and Cursor access to Mobbin's
261. 9:14library. So, instead of making things
262. 9:16up, [music] your AI can reference
263. 9:18patterns that already exist in the real
264. 9:20world, compare how apps design their
265. 9:22paywalls, and pull examples directly
266. 9:25into your workflow. Think of it as
267. 9:27grounding your AI's output in what's
268. 9:30already been shipped.
269. 9:31Not every pattern is worth copying.
270. 9:34Some of the highest converting paywalls
271. 9:36are the ones Jonathan is the most
272. 9:38skeptical of. The same tactics you'll
273. 9:40find in drop shipping stores and
274. 9:42e-commerce websites.
275. 9:44>> So, one paywall that works extremely
276. 9:47well that I really have a hard time
277. 9:50accepting is the spin the wheel paywall.
278. 9:52This is a predetermined spin the wheel
279. 9:55Lottie file that may land on some kind
280. 9:57of discounted percentage. You know, the
281. 9:59user gets to claim it and then cuz
282. 10:02they've just won something, the idea is
283. 10:04that they'll uh be more likely to then
284. 10:06subscribe at that price point. If you're
285. 10:08optimizing for a longer-term business,
286. 10:11that may not be a good paywall for you.
287. 10:13But, if you're optimizing for a weekly
288. 10:15price and you want to maximize revenue,
289. 10:18unfortunately, it works. The issue is
290. 10:20that everyone is also starting to use it
291. 10:22or I'm seeing it used a lot more.
292. 10:24>> When every app uses fake urgency
293. 10:27tactics, users eventually stop believing
294. 10:30it.
295. 10:30>> I think more and more users are getting
296. 10:32comfortable closing paywalls because
297. 10:34they know they're going to get some kind
298. 10:35of offer afterwards.
299. 10:37>> And in early 2026, Apple also started
300. 10:40rejecting paywall patterns that relied
301. 10:42on free trial toggles because some of
302. 10:44these were considered confusing or
303. 10:47misleading.
304. 10:48>> Just remember that this is a long-term
305. 10:49game.
306. 10:50>> Jonathan once worked with an AI
307. 10:52girlfriend app.
308. 10:54>> They had probably the ugliest paywall
309. 10:56I've ever seen and it was really compact
310. 10:58elements, very aggressive squeeze page
311. 11:01kind of paywall. So, that ended up
312. 11:04over-performing pretty much any paywall
313. 11:06that I tried to do.
314. 11:08>> Which means that there is literally no
315. 11:10universal best paywall, only better
316. 11:13experiments.
317. 11:14>> One thing that I learned from Jake and
318. 11:16Nick at Superwall was that design tests
319. 11:19move the needle the most. We ended up
320. 11:21running radically [music] different
321. 11:23designs until you find something that
322. 11:25blows everything out of the water. You
323. 11:27could have a video paywall, you could
324. 11:28have bullet list paywall, a trial
325. 11:30timeline paywall, a long-form [music]
326. 11:32paywall.
327. 11:33>> Test every copy, the pricing, the flow
328. 11:36until something clearly wins. Getting
329. 11:38people to subscribe is one thing.
330. 11:40Building something that people continues
331. 11:42to pay for [music] is much harder.
332. 11:45>> The metrics that I wish more founders
333. 11:47really care the most about is retention
334. 11:49and LTV. I think if retention and LTV
335. 11:52are good, you're building a good product
336. 11:54and you want to build good products so
337. 11:56that the ecosystem is flourishing in
338. 11:58itself. So,
339. 11:59>> maybe the question isn't how do I get
340. 12:01people to [music] pay? It's more of how
341. 12:03could I create something worth paying
342. 12:05for?
343. 12:07Fun fact,
344. 12:07>> [music]
345. 12:07>> it takes less than five screens to
346. 12:09subscribe to ClassPass, but 17 screens
347. 12:12to cancel. Let us know if you'd like us
348. 12:14to dive into cancellation flows or
349. 12:16anything else in the comments below.
350. 12:18Next, watch our breakdown on onboarding
351. 12:20flows, dashboard designs,
352. 12:23and the psychology behind streaks.
353. 12:26I'll see you there.