Case Study: B2B SaaS / Sports Analytics
PickTheOdds had real arbitrage math buried under navigation that broke past six links and a filter panel nobody could parse. In a one-month redesign of the interface, daily active users grew from 50 to 500+.
01: the evidence-backed one
I polled the product’s 700+ member Discord. 68% of users found the filters too slow, and 72% said they’d pay for better ones. I grouped the filters into categories and A/B tested that against the old panel. The grouped version cut the time to select a betting category by 31%.
What I didn’t do: shrink every filter to fit. That would have kept the same flat hierarchy in less space, and the hierarchy was the real problem.
02
Every user runs the same loop all day, from the Odds screen to the matching Game screen. I rebuilt the navigation around making that path as short as possible. This is the most likely driver of the engagement gains: 84% more time on site and 25% higher retention.
Trade-off: paths people used less got less attention in a one-month window. That was deliberate, because the core loop is where the product earns its money.
03
The product had no subscription flow. Instead of adding a separate settings page, I put plan and billing into one central account panel, so paying fit into the same navigation fix happening everywhere else.
Trade-off: more account architecture up front than a bolted-on page, in an engagement that had little time to spare.
Daily active users grew tenfold over the engagement. No single change explains that on its own, but the filter and navigation fixes were the structural changes behind all three metrics moving together. Every wireframe and prototype went back through the same Discord community before any visual design started.
“The new navigation and filters are a game-changer. Engagement and subscriptions have never been higher.”
Ken Bean, Founder, PickTheOdds