Converted to paid users 1.6% of page traffic consistently during 1.5+ years
Parlay Picker is a feature on Dimers.com that helps people build parlays without the usual hassle. Instead of picking each leg one by one, users set filters, like sport, sportsbook, bet type, or number of legs, and instantly get a parlay built for them. The idea was to make parlays faster to create, easier to understand, and backed by real data instead of guesswork.
I designed the full user flow, from choosing filters to seeing the final parlay. I turned feedback and analytics into wireframes, prototypes, and final responsive screens. I made sure the design could handle a lot of live data (odds, probabilities, edges) without overwhelming the user. I worked closely with developers to bring the feature to life and tested designs with real users to make sure it was simple and clear.
Worked directly with: 1 Product Manager (feature planning, prioritization), 2 Frontend Developers (implementation, QA), 1 Data Analyst (impact measurement), and Data Scientists (responsible for building and refining the LLM-powered models that generated Dimers' betting recommendations). Collaboration was agile and highly iterative, with regular stakeholder reviews.
The hardest part was showing a lot of numbers, odds, probability, edge, payout, without making the screen feel like a spreadsheet. I had to find the right balance: clear enough for casual users, but still detailed enough for serious bettors.
We started by looking at user feedback, which showed people found manual parlays frustrating. I sketched early ideas for filters and layouts, then built low-fi prototypes to test the flow. We ran quick usability sessions to see how people interacted with the feature. Each round of feedback led to tweaks, from how filters were displayed to how parlay results were stacked. Once the flow felt smooth, I created high-fidelity designs and worked with developers on the final build.
The goal was simple: help users create strong parlays quickly and with confidence. Instead of guessing which legs to combine, the feature shows them data-backed picks. By keeping the process quick and easy, Parlay Picker aimed to save users time, reduce confusion and decision fatigue, build trust by being transparent about the data, and increase betting activity by making parlays more accessible.
More users placed parlays compared to when they had to build them manually. Engagement increased as users clicked around more with filters, showing they enjoyed exploring options, and session time went up as users tested out multiple parlay builds in one visit.
Analysts confirmed the feature drove more parlay engagement. PMs wanted to keep it simple, so we focused on a clean interface. Developers asked for lightweight designs so the live data wouldn't slow the page, which shaped the final component design.
Parlay Picker showed that a complex product like parlays can be made approachable with the right design. I learned that users want a mix of control and guidance, too much automation feels like you're taking choices away, but too much manual work makes it tiring. In the future, I'd explore personalization, like tailoring parlays to a user's betting style, and adding live odds updates directly into the parlay view.
Mayorga