Case Study: How Product Consulting & User Research Delivered 23% Conversion Growth

When a fintech startup brought me in to help with their product strategy, I did not expect the thing holding back conversion to be something users could have told us in an afternoon. It took three weeks of data work and 20 interviews to find it — and four weeks to fix it.

The startup that had everything except conversions

Linx is one of those startups that just makes sense. They've built a cashback platform where businesses reward customers for sharing purchases socially. The economics are neat: businesses get organic social traffic, customers get cheaper purchases.

The team is strong, the user base was growing fast, and the fundamentals were solid. But within days of joining as their product consultant, we found something that stopped me short.

The data that changed everything

First things first: they needed data foundations. When I arrived there was almost no instrumentation — a common startup problem, and the reason so many teams argue from opinion. We agreed the metrics, KPIs and goals, built the event taxonomy, worked with engineering to implement tracking, and stood up daily and monthly dashboards.

Once we could see, the problem jumped straight out. A large share of users were downloading the app, getting free money credited to their account, and then doing nothing at all. Zero purchases.

Which is a strange thing to watch. Why would anyone ignore free money?

User research revealed what the data couldn't

Data shows you what is happening. It cannot tell you why. That gap is exactly what customer interviews are for.

Working closely with the CPO and their designer, we went straight to the source: 20 customers across three distinct segments.

1. Brand new users who just installed the app
2. Existing users who tried it once and never returned
3. Power users who engage weekly

We built a questionnaire covering their background, why they downloaded the app, what they expected from it, and how it fit their daily routine. The aim was to understand the real motivation behind installing — and behind not coming back.

It took a few iterations to get the questions right. Once they were, the pattern showed up fast.

The moment it clicked

What we found reshaped major parts of the product strategy.

A large share of users simply did not understand how Linx worked. And here is the part that made it actionable: the users who did figure it out became the most engaged customers on the platform. They used it often and told their friends.

So the ceiling wasn't the product. It was comprehension. Fix that, and both conversion and retention would move.

Two paths, one decision

We had two options, each with a clear trade-off:

1. Redesign the core app experience. This addresses the root cause, but working out the exact fix would be slow and uncertain.
2. Build an onboarding experience. This doesn't solve the root cause, but it tests cheaply whether clearer messaging can bridge the gap.

The part that is really just prioritization

Most of the value in product consulting comes down to choosing well between good options. When a startup has limited engineering time, every choice has knock-on effects across the whole company.

We used a plain effort-versus-impact comparison — no need for a heavier process when the team is small and moving fast. The maths was clear: redesigning the core flow was 2+ months of work, while a simple but well-made onboarding was about 2 weeks.

Given how much we wanted to learn quickly, onboarding won. Worst case we lose two weeks. Best case we unlock the growth curve.

Execution at startup speed

1. Week 1 — initial design concepts and the messages we needed to land.
2. Week 2 — design iterations and refinement.
3. Week 3 — development and testing.
4. Week 4 — launch.

Four weeks from decision to live, on a two-week build estimate that held.

The results

Same-day conversion — app install to first purchase — jumped 23% overnight. Then it kept climbing. The general version of what we did is worth reading if you run a mobile app.

For a two-week build, that is an unusually good return, and honestly more than we expected from what amounted to conversion optimization through better explanation.

Why it worked

Three things, none of them clever: reliable data to find where users dropped, real user research to explain why, and the discipline to ship the cheap test rather than the expensive fix.

For the Linx team it was a starting point rather than a finish line. They kept iterating, and kept pushing that number up.

The bigger picture

This case study reinforced something I see in product consulting again and again: the highest-impact fixes usually come from understanding users properly, not from technical sophistication. The comprehension problem was invisible in a feature backlog and obvious in an interview.

It's a particularly sharp lesson for any fintech product. When money is involved, people don't experiment their way to understanding — if they can't tell what will happen to their cash, they simply don't press the button.

Solid user research, honest prioritization, fast execution. That is most of the job.

If you want to see the full version of that sequence, here is how an engagement runs end to end — the ten steps from discovery and instrumentation through research, strategy and iteration.