Rewards that felt personal

The pattern was simple. Someone needed money before payday. They opened Earnipay, withdrew it, and left.

There was nothing wrong with the withdrawal experience itself. The problem was what happened afterwards.

Earnipay knew when someone needed their salary, but we were not giving them many reasons to come back when they did not. The relationship with the product had to be bigger than one transaction.

That led us to rewards.

Rewards were a natural way to give people another reason to return. But there was a problem with the obvious approach.

We wanted the rewards to feel like they were meant for the person looking at them. So the question became:

How do we personalize rewards around what each customer actually cares about?

But to personalize something, we needed to know something about the person. And we didn't.

At the time, Earnipay had no way for customers to understand where their money was going. Transactions existed, but spending behaviour was not being surfaced anywhere.

That meant we could not easily tell the difference between someone who spent most of their money on food and someone who spent most of it on transport.

So before we could make rewards personal, we had to understand spending. That became the next product problem, and it is where Spending Insights came from.

We grouped transactions into categories so customers could see where their money was actually going. Food, transport, bills, shopping, and everything else.

The important part was not the categories themselves. It was the picture they created.

Instead of scrolling through a list of transactions and trying to remember what each payment was for, customers could see the shape of their spending. One clear picture, not a list.

Where people did want the detail, they got to choose how to read it. Some read money as a trend across the month, some want one day next to another. It is one tap between them, so nobody has to learn the version I happened to prefer.

Spending Insights was never meant to be a dashboard people looked at once and forgot about. It gave us something we did not have before: a way to understand customer behaviour.

If someone consistently spent more on food, their rewards could reflect that. If another customer spent more on transport, their rewards could be different. The experience could move from here are some rewards to here are rewards that make sense for you.

That connection was the real idea behind Insights. We were not building analytics for the sake of analytics. We were building the layer of information that made personalization possible.

With spending behaviour available, we could think about rewards differently. They became tied to activities inside the product: customers completed activities, earned points, and spent those points on rewards. Some of those worked like coupons, including food offers.

Once points became part of the experience, progress had to feel tangible. That introduced another part of the system: the leaderboard. It gave customers a way to see how they were doing next to other people.

Your points were not just a number anymore. They were progress, and progress gave people another reason to return.

Activities became points, points became progress, and progress became something useful.

Points for the habit, not for the spend.

A few months after rewards launched, we noticed something unusual. One person was accumulating points much faster than everyone else, and the leaderboard made the difference obvious.

At first it looked like someone had simply worked out how to earn points quickly. So we investigated.

The account was not behaving like a normal customer account. The credentials used to sign up were fake, and the account was engaging in malicious activity on the platform. We removed it and resolved the issue.

What we had built as a gamification feature had become a useful signal for spotting abnormal behaviour. The leaderboard was never designed as a security system. But because it made behaviour visible, it helped us notice something that would otherwise have been harder to see.

The original problem was simple. People came to Earnipay for one thing, their salary. The opportunity was to give them more reasons to stay connected to the product.

Rewards gave us the incentive. Spending Insights gave us the understanding that made those incentives relevant. Together they created a new loop.

Within 6 weeks, non-withdrawal sessions made up 18% of all app opens. The app was becoming more than a place people visited when they needed money.

The interesting part of this project was not any one feature. Rewards without personalization would have been generic. Personalization without spending data was impossible. Spending Insights without a reason to act on the information would have been another screen people visited once.

Together they made a system, and that changed how I think about product design. Sometimes the feature you are asked to build is not the whole solution. You have to understand what role it plays in the larger product experience.

For Earnipay, the goal was never to make people spend more time in the app. It was to give them more reasons to come back.