Online Ordering

I was the lead product designer for the consumer ordering experience and the operator-facing configuration tools during my time at Cents. I owned both sides of the system — what customers see when they place an order, and what operators configure on the back end to make it work for their business. I worked closely with a PM and a few engineers throughout.

The work happened in layers over three years, each iteration removing friction and reordering the logic to match how users actually think about placing an order.

The first major change was moving service selection to the first step. Users needed to know what they were ordering before making any decisions about timing. Everything downstream of that shifted accordingly.

I collapsed the three scheduling screens — pickup time, delivery time, and recurring order — into a single page with smart defaults. Instead of asking users to make three separate decisions, the flow presented sensible defaults they could adjust. The recurring order prompt was removed as a popup entirely and became a simple toggle, with any applicable discount shown inline.

Roughly 40% of users complete their order without ever touching the time selector, meaning the default works for them out of the box.

The most recent and significant change was introducing a cart page and multi-service selection. This wasn't adding a step — it was reorganizing existing information into a mental model users already understood from food delivery and e-commerce. Information that had lived across multiple screens was consolidated into a familiar cart experience.

Multi-service selection was also a meaningful product expansion, and part of a broader effort to redesign the service model for flexibility and scale. Because Cents is a white-label platform, operators structure their businesses in different ways. Building a system that could accommodate more business models means more operators can offer the high-value services that drive the most revenue — Pickup & Delivery orders are the highest-grossing order type, for operators and for Cents. Enabling customers to add multiple services in a single order drives more revenue per transaction, which compounds across a growing operator base.

This overhaul was validated through AI-moderated user testing sessions via Strella before launch. All 7 participants easily completed every task.

Because the Cents B2C experience is a white-label platform, no two operators run their business the same way. Some offer pickup and delivery only. Some offer in-store services. Some have complex pricing tiers, custom service categories, and specific scheduling windows. The consumer ordering experience is only as good as the configuration layer behind it.

I owned and redesigned the operator-facing configuration tools in parallel with the consumer experience — building flexibility into the system so operators could structure their services the way their business actually works, rather than forcing them into a rigid template. This also included a vision for scaling the service model to accommodate more business types over time, which directly supports Cents's growth as a platform.

Designing both sides of the system meant every consumer-facing decision had to account for operator configurability. That constraint made the design work harder and more interesting — the kind of systems thinking that doesn't show up in a single screen.

The biggest thing I'd do differently is push for instrumentation earlier. Making decisions without conversion data is manageable — good UX principles and customer feedback go a long way — but having numbers would have made every argument for change faster and easier to win. Getting tracking in place early is something I'd advocate for on day one at any new company.

The roadmap ahead is focused on depth and retention: