How efficient navigation solved our biggest reordering obstacle

Overview

On a B2B platform built for bar owners, restaurant managers, and shop owners placing recurring orders, every extra tap is friction. These are people ordering between shifts, often on mobile, with limited time and a clear goal: get in, reorder, get out. The faster they can move between the product catalogue, their order history, and their loyalty points, the better the experience. And for a long time, that journey was more cumbersome than it needed to be.

The hypothesis

If key sections were made more visible and directly accessible in the mobile navigation, users would move between the platform's core areas more efficiently. We expected to see an increase in visits to Orders and Club, and a reduction in the friction caused by repeatedly opening and closing the hamburger menu.

What we designed and tested

I designed a bottom navigation bar for mobile, giving direct access to Products, Orders, eazleClub, Account, and Cart. To run a clean test, we also removed the Cart icon from the header bar and kept the hamburger menu visible, specifically to see whether users would gravitate toward the more obvious navigation option even when the familiar one was still there.

The A/B test ran for six weeks across all mobile sessions in Myanmar, with a 50/50 traffic split.

What the results told us

The primary goal was achieved. Visits to the Orders page increased by 27.9% and Club visits grew by 3.4%. The bottom bar was used in 64.3% of all sessions, and hamburger menu usage dropped by 22.1%, which told us users preferred the more accessible navigation when it was available.

But the test also introduced friction we hadn't fully anticipated. Users expected the Cart icon to stay in the header bar, where they'd always found it. Moving it to the bottom bar led to a spike in Account page visits driven largely by mis-clicks from users aiming for the Cart. The reduced viewport also made product browsing slightly harder.

Iterating before shipping

The recommendation from the test was clear: don't implement as tested, iterate first. That's exactly what we did.

The core insight was that the bottom bar needed to be purely a navigation surface, not a home for transactional actions like the Cart. So in the iterated version, the Cart moved back to the header bar where users expected it, freeing up space in the bottom bar and reducing mis-clicks. The navigation bar was also set to disappear on scroll, giving users back the viewport space they needed while browsing the product list. The result was a cleaner separation between navigation and actions: top bar for session-critical tasks, bottom bar for moving between sections.

What I'm most proud of

We didn't just implement a popular UI pattern because it felt right. We traced a real user behaviour from behavioral data, turned it into a testable hypothesis, ran a proper experiment, read the results honestly including the parts that didn't go our way, and used that to shape a better solution. We shipped the navigation bar because six weeks of data proved it was the right solution, not because it looked good in a mockup.

What I learned

Familiar patterns can generate certain expectations, and moving the Cart caused confusion among our users, which we did not anticipate. That doesn't mean we should push back on change, but we need to be aware of what conventions we are disrupting and why. Testing the Cart in the bottom bar was a valuable decision in hindsight, because it gave us certainty about the consequences of disrupting too much too soon, and luckily we managed to adapt to that before we shipped.