What the best subscription apps get right about paywalls

Summary: Top subscription apps design paywalls as part of the full user journey. Growth expert Rosie Hoggmascall shares tactics for pricing by stage, personalizing flows, and measuring quality through refund and churn rates — with real examples from 20+ apps.

For the last six months, I’ve been absolutely devouring Rosie Hoggmascall’s newsletter, Growth Dives. In this weekly newsletter, Rosie provides annotated breakdowns of Saas products to help people get ideas about what to build. I love that Rosie frequently returns to analyzing the paywalls and pricing strategies of various subscription apps sharing actionable insights.

I’ve found that content surrounding paywalls tends to be quite focused on tips and tricks. Whilst that can be very helpful, what I loved about Rosie’s approach, in particular, is that she clearly considers these as part of the broader growth strategy. Rosie has both analyzed and worked with 20+ subscription businesses, so who could be a better expert on the subject? Luckily, Rosie was happy to oblige and share what she’s learned about setting up your paywall for success.

First things first: let me introduce you to Rosie. 

Rosie is a Product Growth expert specialising in PLG, UX, and monetisation. She’s now a Fractional Head of Growth, and has worked with incredible subscription products like Blinkist, Miro, Thriva, Fyxer.ai and more.

In just over a year, Rosie amassed almost 15K followers on Medium with her intricate breakdowns of well-known apps like Spotify, Duolingo, Figma, and Strava. This led Rosie to start her weekly newsletter, Growth Dives, where she breaks down strategies so others can learn from them. 

We discussed common mistakes in paywalls and how to avoid these, web-to-app versus app payment flows, metrics to consider, and much more. I can honestly say that I learned so much from this conversation, and I have no doubt you will, too. Let’s take a journey into Rosie’s brain. 

Key takeaways

• Paywalls work best when integrated into the full user journey.
• Track refund, cancellation, and complaint rates to assess quality.
• Personalize paywalls and optimize their timing.
• Treat in-app and web-to-app flows as distinct systems.
• Post-purchase UX affects retention and lifetime value.

What are common mistakes subscription apps make when it comes to paywalls?

When it comes to paywalls and pricing mistakes, there are two groups: early-stage and later-stage apps. 

In the early stages, it’s more a case of the pricing strategy going wrong. Rosie often sees new startups waiting too long to price and being scared to build in a subscription offering. When they do finally make this leap, they frequently price their app too cheap and leave money on the table. They essentially undervalue their own brand, first in terms of not charging, and then in charging too little. In Rosie’s experience, it’s easier to price high and then lower. 

The example she gave is to imagine a group willing to pay £20 per year for your app and another group willing to pay £100 per year. Most people will worry that you lose the £20 per year group if you price at £100 per year. But actually, if you price at £20 per year, you miss out on revenue from those users who are willing to pay more. If you start with £100 per year and then offer discounts later in the flow, you can get both groups closer to what they are willing to pay. So, both her learnings are that in the early stages, it’s much more about optimizing your pricing go-to-market strategy than your paywall.

Later-stage startups often work out their pricing strategy and spend a long time optimizing. Here, the main mistake she sees is putting business goals before user goals, often resulting in frustrating experiences or dark UX. This could look like focusing on CRO without a measure of quality, e.g. retention.

I think this is a key lesson for all stages: focusing only on conversion rate as a measure of your success will hold you back without a check metric of quality. For instance, optimising for trial opt in without measuring trial conversion, or measuring for new subscribers without measuring cancellation or churn rate. Dark tactics may result in short-term gains, but it will leave your users unhappy and frustrated with your purchase flow.

Some examples of dark-leaning apps are MyFitnessPal, a calorie-tracking app, and some could argue Flo, a hormonal health-tracking app. She’s also seen this issue with game apps, where accidental purchases seem to occur more often. 

Rosie did an analysis on MyFitnessPal (How NOT to activate users: lessons from MyFitnessPal) and found that the actual paywall wasn’t terrible; in fact, they had a lot of the best practices in play:

Credit: UX Collective, “How NOT to activate users: lessons from MyFitnessPal” by Rosie Hoggmascall

However, in the pre-paywall flow, MyFitnessPal promised personalisation (but failed to deliver) followed by two dense data privacy screens designed to make the user zone out of the T&Cs instead of help them understand.

Credit: UX Collective, “How NOT to activate users: lessons from MyFitnessPal” by Rosie Hoggmascall

At that point, they may have seen conversion rate gains (we don’t know this for sure) purely because people were rapidly clicking through in an attempt to escape the giant wall of text — can you blame them?

The lesson here is to not just consider your paywall (which MyFitnessPal do well) but also the pre-purchase experience. Asking: what does the user need to build trust? What unanswered questions do they have? 

Any paywall flow where the user just taps to get through it a sort of ‘let me just get this over with’ feeling, or walks away feeling overwhelmed or tired is a paywall that may have long-term adverse effects in terms of higher cancellations and refund rates. 

A design that aimed to reduce this was the Honest Paywall, first popularized by Blinkist, a learning app. This marked a shift into apps testing more transparent paywalls. Note that it won’t work for everyone, but it’s an interesting case study to learn from. Blinkist redesigned its paywall to center on three key components:

• A clear timeline
• The features included
• A guarantee that they would remind users of their trial

The result? It increased their conversion rate by 23% and dropped complaints by 55%. Those numbers do not lie. Often, in growth, we feel the temptation to ‘spin’ things or provide our best face to customers, but the Honest Paywall is a reminder that trust is paramount to a brand’s relationship with its customers. It’s hard to build and easy to break, but boy, is it worth it.

Credit: UX Planet, “How Solving Our Biggest Customer Complaint at Blinkist Led to a 23% Increase in Conversion” by Jaycee Day

What metrics should you use to effectively optimize your paywall?

Okay, we’ve gone over what you’re likely doing wrong when it comes to paywalls, but now we need to consider what you should be doing instead. Namely, it’s time to talk metrics. If conversion rates are misleading at times, which metrics can provide a better idea of performance?

In Rosie’s experience, there are a few additional metrics to keep an eye on:

• Refund rates 
• Cancelation rates
• Complaint rates

With longer trials (such as 30-day trials), trial conversion rates can be tricky, as your tests take much longer to show results. Instead, Rosie recommends looking at the frequency distribution or average period cancelations that occur to ensure you capture the most accurate metrics. 

She’s found that 20-30% of users tend to cancel in the first three hours, so even measuring beyond that point can ensure your paywall tactics aren’t bringing in just low-quality users.

It’s all about going beyond the trial conversion rate and getting a more accurate performance measure. We see this with Blinkist, too, who looked at complaint rates as a measure of the success of their honest paywall.

So, the best apps use more holistic metrics to measure success, but what else do they do differently?

How do the best subscription apps approach pricing differently?

I was also curious about what Rosie saw the best of the best doing well. She mentioned two key things: contextualizing paywalls and not treating paywalls in isolation.

How often your user sees your paywall will impact their conversion rate. It might sound obvious, but you’d be surprised how rarely brands consider this in practice. After just one interaction with the prospect of a paywall, some customers might not be ready to convert. Instead, they’ll probably want to see it a few times before making that leap. 

In Rosie’s experience, the opportunity in this differs depending on the maturity level:

• Early-stage startups could show their paywall more
• Mid-stage startups are showing their paywall more frequently but could further personalize it 
• Later-stage startups are doing both: showing it enough and modifying it, but sometimes do show it too much, to the extent that it feels like spam

So what’s the right answer? Ideally, you want to balance the two: show it enough and personalize it. To understand where the opportunity lies, Rosie tends to monitor two metrics closely: the number of times the paywall has been viewed and the paywall-to-conversion rate. That way, she can work out where the opportunity lies.

The best apps think about what you were doing before you clicked through to the paywall and adjust it based on that. She said, “The best paywalls feel like they speak to you,” which brings in that second part of not treating it in isolation from the rest.

Rosie once tested adding three screens before the Paywall that provided additional context. She expected that lower friction would improve conversion but wanted to test it to be sure. The result was that adding that three-slide carousel actually boosted the conversion rate: it gave users an understanding of why they should pay. Roughly she saw the following results

• A: welcome screen, onboarding, home, paywall – 2% trial opt in
• B: welcome screen, paywall, onboarding, home – 8% trial opt in
• C: welcome screen, new three part carousel, paywall, onboarding, home – 15% trial opt in

This reminded me of an article that had popped up earlier that day in my inbox, about how SmartTales, a children’s app, had improved their paywall conversion rate by 86%… without actually changing their paywall. Much of the impact came from focusing more on their Jobs to be Done in their ads and ad store listings. So much happens both before and after the paywall.

That period after the paywall is more important than you realize. This is where you’ll see a change if you measure the secondary metrics we discussed. Rosie gave the examples of Strava and Duolingo. 

Strava, the fitness app, has different visuals after subscribing and reminds users of the value of Strava:

Credit: UX Collective “Strava’s bumpy road to monetisation” by Rosie Hogmascall

It feels weird to do this post-trial sign-up… they signed up already? Don’t they understand the value already? Not necessarily! 

Rosie has found that when you survey people after they’ve subscribed, they’ll fail to tell you the app’s primary value. As an ADHDer who rushes through way too much and likes trying out every app on the planet, I’ve definitely failed that survey more than once — oops.

I call this ‘aftercare’ mainly because most apps miss that care part. You’ve just committed to trying out an app, and in most cases, the store equivalent of shoving you out the door once you’ve paid is the aftercare part. Even a single screen of information can go a long way at this point. In the aftercare you want to:

• Validate and celebrate their decision to sign up 
• Share what features they have access to 
• Encourage them to start using the app

If we look at the Duolingo pre-purchase flow, we see them already building in this value and validation upfront, like Rosie’s earlier example (with the three-page carousel):

Credit: UX Collective “How Duolingo drives subscription conversion” by Rosie Hoggmascall

The minute you subscribe, the owl has power, any pokemon fans will recognise this little pikachu like moment:

Credit: UX Collective “How Duolingo drives subscription conversion” by Rosie Hoggmascall

That is one ecstatically happy and powerful owl. Then, post-purchase you get the after-care we talked about earlier. Duolingo walks you through your five benefits again (one of which is just a new icon): 

Credit: UX Collective “How Duolingo drives subscription conversion” by Rosie Hoggmascall

Now we’ve talked a lot about paywalls, but with the rise in web-to-app thanks to clearer tracking, I felt it would be a missed opportunity not to ask Rosie about the differences between the two payment opportunities, and they are really different.

What are the key differences between optimizing a website pricing page vs. an in-app paywall?

Rosie has had her fair share of apps thinking they can just copy and paste their mobile setup over to web. A few clicks and pages, and better margins are yours! Let’s just say, it is not that simple, not by a longshot. It’s far too easy to underestimate web-to-app funnels. 

For web-to-app, you need a different tech stack, a different monetization funnel, a different activation journey (they’ll probably still need to download the app), and a different way of attributing. Even how you offer and send discounts is different, meaning you might need to adjust your CRM strategy based on where you are sending them. Marcus Burke, who used to work at Blinkist, even told me they have two different ad accounts: one for the web funnel and one for the app funnel. 

You get the point; it’s different. 

What that also means is there’s the risk of tech dept, attribution debt and design debt. The extra flow adds complexity, so you have to weigh up the real pros and cons ahead of time.

Now for the similarities, and yes, there are some. The key similarity is that the majority are on mobile. This might sound obvious, but for some reason, the minute app teams think of web, they think in terms of desktop when people are actually still on mobile. Just think of how often you do things on your mobile browser rather than heading over to a computer.

Another consideration is that there is less regulation on the web, which can give you more freedom in your pricing. In the app, if you put an annual subscription first and don’t put the total price (instead listing price per month or day), your paywall will get rejected, whereas you can do what you want on the web. It’s a wild west for pricing pages. However, this should not be an excuse to use dark UX — there’s no excuse for that. As great as pricing psychology can be, keep in mind those metrics we talked about earlier: refund and cancelation rates. 

A final point we talked about is the activation journey. This is something I’ve personally seen an app struggle with, as people would pay for a trial or even a full year upfront… only to not download the app. I’m not talking about a few individuals here; 30-40% of users didn’t download the app in the first week. This was mind blowing to me, but according to Rosie, it is quite common. She highlights how key it is to directly push them to the App Store and ensure there is a seamless way to immediately use your account, such as through a magic link. 

We’ve all had that moment where we want to cancel an app and frantically search in the app or check our subscription, only to discover that we have to go to their website to cancel — urgh. Headspace, the meditation app, is an excellent example of how to approach this situation gracefully. They share really clear instructions on how to cancel:

Screenshot from headspace’s cancellation flow

This mindful product has a really clear setup on how to cancel your subscription in its flows.

A quick-fire round of questions on paywalls

I couldn’t resist ending the interview with a bit of fun. I  designed (potentially the first ever) a paywall quick-fire round of questions. In all honesty, it felt like hosting a gameshow.

In just 2-3 minutes, Rosie covered seven questions on the topic. I was kind enough to allow her to respond “it depends” where relevant. Here are the results:

1. Highest plan first or lowest first? Lead with annual plan and ensure there is only one choice to make
2. Show all options at once or progressive disclosure? Progressive disclosure
3. Comparison tables: effective or not? For complicated high price point (e.g. therapy) products, yes. For simple products, usually not
4. Monthly pricing as default or annual pricing as default? Annual
5. Highlight the “Most Popular” plan or let users decide? Most popular. It’s better than highlighting the “cheapest plan” I’ve found through testing
6. Exit intent popups with discounts: effective or annoying? It depends on the product and the consideration, but it is always worth testing 
7. Dark mode pricing page: trendy or unnecessary? Depends on the target audience, with a younger audience, it’s necessary. They are usually used more by Gen Z  and millennials

As you can see, a lot falls into that dreaded “that depends” category. This is part of what makes paywalls hard. You really need to work out what works for you.

How do you work out what your successful paywall looks like?

As much as these insights help you consider your broader pricing strategy and paywall, I wanted to make sure we ended with how you can take this and start improving your paywall. 

The two metrics we discussed earlier — the number of times the paywall is viewed and the paywall conversion rate — will help you determine where to focus.

From there, Rosie recommends creating a long list of all the variables and elements of your paywall strategy — not just the header, image, and text but also where people see it and what the pre- and post-screen is. Write down all the ideas and assumptions the team has about those areas, e.g., there are too many distractions, there’s not enough information, or we don’t show the paywall often enough. Use this to mock up a few alternative paywall screens or flows.

Then you can get on calls with customers. Yep, I’m suggesting user interviews again! You need to truly understand your customers and adapt to them. For example, if they are risk averse, you’ll probably want a longer flow with more social proof and reassurance. Rosie suggests arranging a call with 5-10 customers. Start these conversations by finding out what other app subscriptions they have and more about their subscriptions, e.g., what led them to subscribe and whether they opted for a monthly or annual subscription.

Then, spend the last 10-15 minutes showing them the mockups you created. Go through them one by one and see what their reactions are. Ask them what they notice about them. Whenever I do this, I’m amazed at how things differ between our perception and theirs, including words that they don’t like or struggling to find key information that we think is obvious. It’s always so insightful.

This, combined with the previous advice, will help you approach your paywall better and set it up for success. Think about it in terms of your broader growth strategy rather than a single screen in your app. Your paywall isn’t just about getting a customer to pay; it’s about setting yourself up to keep them as customers in the long term. Take this approach into your paywalls, and you’re geared for success.