Sell the job, not the features: how JTBD paywalls boosted conversion by 169%

Léa Samrani just gets app growth. She’s worked in product roles at apps Bumble, Uptime, and Passion, all of which have millions of users. (Yeah, I’m fangirling a little over here.)

So I had to ask her: where has she seen the biggest return-on-investment (ROI) in terms of improving app monetization?

Spoiler alert: it wasn’t new features, it wasn’t new designs or focusing on improving retention. 

It was the unsung hero: Jobs to be Done (JTBD). Aka, understanding what your end user is actually trying to achieve. 

Since then, Léa has advised over 45+ apps as a freelance Product and Growth advisor, and the biggest ROI consistently comes from focusing on JTBD. As she puts it: “If we could publish the ROI number, everyone would spend more time on user interviews.”

JTBD success stories

But not everyone wants to spend time speaking to users and understanding their JTBD. So Léa’s developed a strategic approach to JTBD research — one that immediately feeds into paywall and onboarding optimizations, driving quick wins and justifying deeper investment in customer insights. 

The results speak for themselves. Here are just a few she’s proud of:

1. Kids app, Smart Tales: 72% increase in free-to-paid conversion rate
2. Teenager app: 322% increase in ARPU 
3. Community learning app: 169% increase in free-to-paid conversion rate

Those are some impressive figures. But what’s most interesting? All these apps were in the same education category, yet each uncovered a completely different JTBD. Proof that you can’t assume, and you definitely can’t copy-paste a JTBD from your competitors.

Léa says this is what makes JTBD-led optimizations both the easiest and hardest trick in the book: there’s plenty of content out there on JTBD, but doing the customer interviews — and implementing the insights well — is the hard part. You can’t just copy your competitors. You have to go out and deeply understand your users.

And here’s the other catch: it’s never just one test. The teenager app involved around 10 experiments, while the community learning one was about 20! This isn’t about finding a single silver bullet; it’s about building a strategic roadmap to sustainably improve your free-to-paid conversion.

I sat down with Léa to unpack the success of SmartTales, and walk through her step-by-step approach to using JTBD to optimise your paywall.

✨ Pro tip
With RevenueCat’s Paywalls v2 recently released, testing your paywall is quicker than ever — making it even more worthwhile to invest in upfront research and set your future experiments up for success.

Most founders don’t want to focus on JTBD — why?

Imagine you’re a founder (if you are one, no imagination needed). You’ve poured countless hours, sweat, and tears into building your startup from the ground up. Every inch of it reflects the problem you’ve seen time and time again — maybe even struggled with yourself. You’re immensely proud of what you’ve built.

Then, a team member or external advisor (I’ve played this bad cop role before) says:

“You need to focus less on what you think is the most important problem. Take a step back, figure out what your customers’ real problems are, then talk about those problems.”

*Cue internal scream*

It feels so wrong. You’ve created something amazing to solve a problem, and now you’re being told to abandon that and focus on ‘selling techniques.’ No thanks!

But JTBD is far from a ‘selling technique’. The reality is that good products often struggle to communicate the problem they solve.

Léa and I have both faced pushback on using the JTBD framework, but the truth is, the person suggesting this isn’t trying to undermine the original reason for your app (even though it still hurts). They’re trying to help you take your business to the next level by taking a step away from your first-hand experience, and focusing on the highest-paying, best-converting end user. 

However, that’s hard to take on board when the next advice includes:

“By the way, it’s going to take hours and hours of user interviews. Have fun!”

*Cue second internal scream*

And there you have it — the reason most founders shy away from JTBD research. There’s constant pressure to move quickly, grow fast, so the idea of slowing down and learning to conduct interviews doesn’t exactly sound appealing.

Still not sold?

Think of it this way: if it takes 30–50 hours to conduct this research, but you see results like the case studies in the intro (72%–169% increase in free-to-paid conversion, 322% increase in ARPU) — would you consider that a positive ROI? 

Almost certainly, because that value will ripple through the rest of your funnel. So, how can you achieve those results?

Step-by-step JTBD workflow for subscription app paywalls 

Let’s dive into when and how to use Léa’s approach, and walk through the case of Smart Tales, an educational app for kids. (We’ll touch on the other two examples, but keep those anonymous for the brands involved.)

Unlike our hypothetical founder, the Smart Tales founders were ready and willing to put in the work from the start. They met Léa through the SYSTM growth program, where she was their coach, and they were rewarded for their hard work with a 72% increase in free-to-paid conversion rate across a series of experiments — incredible results. 

Here’s how they did it, and how you can too. 

1. Identify signals you need to focus on JTBD

The first step is the shortest — you need to notice the problem. 

Smart Tales has a genuinely great product, and their retention figures reflect that. Even now, with over 11,000 reviews, they maintain a 4.6 rating on the Apple App Store. However, at the time, the front of the funnel was lagging. They were focused on highlighting the quality of their lessons and content, but they were missing the core JTBD. While their app helped kids practice math, science, and reading, that wasn’t the real problem they needed to solve.

In general, strong retention paired with low conversion rates (and often low click-through rates) is a key indicator of a JTBD issue.

2. Practice to conduct strong user interviews

Léa explains that customer care logs, reviews, and user tests are all great, but they don’t allow you to dive deeply into the customer’s job and the problem you’re trying to solve. These sources can be too leading — often feature-focused, rather than asking the essential questions you need to uncover.

Conducting user interviews well is the hardest part. But Smart Tales, once again, didn’t deter. They practiced before conducting some of the best user interviews Léa had ever seen. She strongly advises taking the time to practice your user interviews first if you don’t have experience.

How? Ideally, don’t talk to friends and family, who you (hopefully) know too well. Instead, try interviewing colleagues or acquaintances at other brands about a completely unrelated product.

Léa once played the interview guinea pig herself, allowing a client to interview her about her living room table. What JTBD was that table serving her? It wasn’t even related to subscription apps, but that’s the key: it needs to be something you know nothing about to avoid falling into assumption mode.

What had seemed like the most standard, run-of-the-mill purchase ever for Léa ended up revealing how much that table meant to her and what it symbolized in terms of buying her first home. She’ll never look at that table the same way again!

2.5. Write your questions (and check them)

I’ve walked through JTBD questions to understand what drives customers to pay before, but I couldn’t resist asking Léa her favorite questions (yes, I’m a JTBD nerd, what do you want from me?). Her favorite question was:

What do you think the app was going to enable you to do? 

Or phrased differently: What did you think you could do with this product? You’re trying to understand: what did they expect to feel, and why did they buy it? How you phrase these questions is key.

Léa once asked a customer about why they purchased a VPN system, their response? “Err… security?” Oops, a bit too literal. While ‘security’ might be the surface-level reason, you need to uncover why they wanted security. What was that security going to enable them to do? For example, peace of mind that their data is secure.

If interviewees struggle to answer, Léa suggests asking about their buying journey — the moment they decided to invest in your app. Ask plenty of contextual questions about when and where the decision took place, as this will help them recall what they were worried about or hoping for.

Reading this, you can see why conducting these interviews and reading those signals is a bit of an art form. It takes practice. So, just like Smart Tales, take the time to prepare—not only your questions but yourself for conducting strong interviews. If you’re worried, Léa suggests conducting the interview with someone else, as they can help ensure you’re diving deep enough or take notes so you can focus entirely on the interview.

3. Examine your user interviews to understand your JTBD

It can be tempting after an interview to throw all your notes into an AI tool and ask it to tell you your JTBD. Please, please don’t.

I was thrilled to see that Léa, like me, takes the ‘old-fashioned’ approach of discussing JTBD, walking through findings and insights together. While AI is a great tool for some types of analysis, there are many subtleties in user interviews that AI can miss, as well as broader context about your app that is key. For example, the right JTBD isn’t necessarily the one mentioned the most, but the one you can solve better than your competitors.

What Smart Tales identified as their biggest JTBD was unexpected. It wasn’t teaching kids math, science, and reading. It was helping parents deal with that dreaded screen switch-off moment.

Parents (and cool aunties like me) know this moment all too well.

For example, I offered to take two of my cousins’ kids to the park the other day. They were all quietly sitting on their various iPads, reluctant to come. I thought I had all the convincing power on my side: the dog (that they love) was coming, there was a playground, even hot chocolate and shortbread (I’m not above bribing them). But nope, not good enough.

While we eventually got the eldest ready to go, all hell broke loose with the youngest. He was perfectly happy behind his screen and not happy to be dragged out. Cue crying and screaming. But don’t worry, once we got to the park, he was overjoyed, racing around having the best time ever. The journey there, though? Let’s just say it was rough.

This is the problem (slightly ironically) that Smart Tales solves, they learned from their interviews. The app ensures kids spend non-addictive time behind their screens, so when switch-off time comes, it isn’t the end of the world. And even if the switch-off time isn’t here yet, they’re still spending positive, enriching time on their devices.

If we try to phrase this as the JTBD, it’s:

When my kid is using a screen
I want a way to ensure they can learn something
So that they are not mindlessly scrolling and addicted

The problem they were solving was screen addiction. With this understanding, it was time to start improving their paywall performance.

Why you can’t copy your competitor’s JTBD

As an aside, let’s dip into the other app examples. While all the cases Léa shared were education apps, they each had a completely different JTBD. 

For the second case, the app audience was slightly older (teenagers), and the JTBD was helping them use the app to pass their exams, giving them confidence that by using the app, they could achieve the grades they wanted and get into their dream universities. For the community learning app, the JTBD came down not to learning, but to a feeling of belonging in a community. 

Even with very similar apps and JTBDs, it’s the nuances in your language and showing how you’ll help users achieve their JTBD that set you apart.

4. Improving the paywall starts before the paywall

Often when we think of paywall experiments and optimizations, we focus solely on the paywall itself. While you can definitely use your JTBD to improve the paywall, Léa finds the most significant impact often happens earlier in the process.

Use the start of the journey to test, validate, and refine your JTBD (there’s a whole guide on using these methods here). Léa explains that this early stage is where you see the biggest volume of users — especially for startups — so you can make a large initial impact, assuming your retention is already strong.

From there, begin thinking about onboarding: how will that process help users achieve their JTBD?

For Smart Tales and the app for teenagers, both apps initially directed users straight to the paywall with little to no context. Considering the JTBD, the team tested adding positive friction through onboarding. This looked like introducing helpful context and questions, to guide users understanding how the app will help them achieve their JTBD.

This doesn’t mean adding questions you want answered, like “Where did you hear about us?” or “How old are you?”. That’s just negative friction.

Instead, one thing Léa loves to test is integrating user reviews that reflect the core JTBD, then subtly asking the user, ‘Can you relate?’ It builds trust and signals that the app understands them, especially when this messaging stays consistent throughout the journey. Smart Tales does exactly this in their onboarding:

Another strong example comes from Headway, the book summary app. They clearly understand their audience’s core problem: not having enough time to read full books.

You can even test a similar approach directly on the paywall. One of RevenueCat’s Paywall v2 templates includes space to showcase a JTBD-focused review, helping reinforce value right at the decision point.

For the community learning app, one of the most impactful experiments was changing the timing of the paywall. They discovered the aha! moment came after six lessons — enough to show users they could consistently learn and rely on the app to structure their progress. Even though more users saw the paywall earlier, placing it after those six lessons drove better results. The takeaway? Timing matters, and the ideal moment to show the paywall is when users start to feel their JTBD is being met.

This consistent focus on JTBD across the journey can make testing more complex. Léa pointed out it’s rarely one big test that drives a lift — it’s a series of thoughtful experiments, with consistent language and messaging that compounds over time. 

So what kind of tests worked best? A mix. Léa found combining larger experiments (like adding positive friction to onboarding) with smaller ones (like tweaking App Store copy and screenshots) delivered the strongest results.

5. Begin to optimize your paywall directly

From there, use those learnings around copy, social proof, and more to optimize the paywall.

The community learning app ran paywall tests changing the copy on the paywall from features-focused to JTBD-focused. Too many paywalls simply list out endless features, rather than exploring the value behind those features and explaining what the app will actually help the user achieve. Strong copywriting on your paywall goes a long way.

An example of how not to do this, in my opinion, is Uxcel Go, the UX learning app (sticking with the education category!). If you look at their paywall, it’s entirely feature-focused:

The closest Uxcel Go comes to a JTBD is “Invest in your design career today”, but even that feels vague and surface-level. Why do I want to invest in it? What do I want to achieve?

While Smart Tales’ original headline, “Accelerate your kid’s growth”, was closer to a JTBD, but it still wasn’t the core JTBD that emerged from their research.

They were consistent in their messaging throughout the onboarding:

Now, this was the result of multiple tests, but one thing that really stood out to me was how much less copy there is in the final version. The onboarding is doing the heavy lifting in communicating the value, which allows the paywall to be simpler and more focused. And it works because it clearly centers on the JTBD and what matters to the parents, rather than listing out features. Another interesting shift? The visual. The original was very functional; the final version focused on emotion.

JTBD research and your pricing and packaging

Sometimes, JTBD research even shows that you may need to fundamentally rethink your pricing and packaging. In Léa’s experience, adding a package is usually an easier sell internally, but changing or removing one can be trickier. That’s why she recommends starting with a lower-friction experiment — something that’s easier to get buy-in for and can show results quickly — then tackling harder, longer experiments. 

There are two key things to consider:

1. You may be underpricing. If so, Léa recommends testing a higher price earlier on. The beauty of JTBD is that you can often charge more without changing the product, because you’re finally resonating with the audience in the right way.
   
2. Not every test will be a win, and that’s okay. Both Léa and I have seen JTBD-led experiments perform better than average, but that doesn’t mean you’ll nail it on the first try. The key is not to test too small. A test that returns no significant result can be worse than one that shows a slight drop, as it tells you less.

Either way, keep testing and keep researching. Remember how many experiments it took the examples to get there. 

Using JTBD to optimize your paywalls 

Reading through this, you might be tempted to think: well, that sounds simple enough. And in theory, it is. As Léa said, it’s the hardest and easiest thing all at once. The hard part usually isn’t the experiments themselves, it’s the research. That’s why Léa zooms in so sharply on user interviews. Her approach: do one form of JTBD research well.

When Léa talks about Smart Tales, you can hear the admiration in her voice: they truly committed to the process, staying open-minded and doing the hard work. 

So when you’re halfway through your stack of interviews, think about Smart Tales — they recently hit $1M in ARR, a milestone that only 0.5% of apps ever reach. In the list of reasons they credit for success, they include: ​​

• “Tens of customer interviews to truly understand the needs of parents and children.” 
• “Hundreds of A/B tests to optimize every aspect of the app’s performance.”

If you want to optimize your paywall using JTBD, it doesn’t start at the paywall. Start by preparing for strong interviews, analyze them deeply, then work your way back from the front of the funnel — testing and refining as you go. 

Simple enough, right?