Bonus Bowls: A loyalty platform for a Fortune 500 pet brand

The client came to us with a problem: their products have the best margins but the worst stickiness. Pet parents pick brands on price, so loyalty was hard to build, and that was quietly weakening the vet-customer relationship too. A rewards program is the obvious fix, but only if pet parents trust it enough to bother.

That’s the gap Bonus Bowls closes. The whole design rests on one promise: the receipt part just works, which is exactly where competitors fall down.

So the real question was:

How do we build a rewards platform that pet parents actually trust, and make rewards actually feel rewarding?

We run every project runs on the same five phases of design thinking. Here's how each one played out on Bonus Bowls:

We needed to understand how pet parents behave: where they buy pet food, how often, what they do with receipts, and what would make a rewards program worth their time. I put together a 5-minute survey targeting Canadian pet parents.

Key Takeaways

85%open to the program — the demand proof and the reason the project existsKey finding

66%want cash, not merchandise

50%don't keep receipts

89%smartphone-comfortable

Top brands mentioned

Top loyalty apps mentioned

Buying habits & audience

30%buy on price

47%spend about $100 a month

40%buy once a month

80%are 20 to 40 years old

Full reportSee the full survey results

Great news: people are open to the program.
Beyond that, we found that pet parents shop monthly (suggesting a similar usage cadence for our platform), the audience is young, tech-savvy, and price-driven, and we've identified their favorite brands and platforms to enable a competitive analysis.

Competitive Analysis

After reading through the survey responses, it was time to investigate the pet brands that came up most often, along with some general loyalty apps. We needed to understand what was already out there so we could pick the best features and avoid what was causing frustration.

To accomplish this research I signed myself up to the programs, downloaded the apps and used them, read numerous Google and App Store reviews, and joined Reddit forums.

Every program we looked at got one half right and dropped the other, so how might we deliver both: reliable, universal approvals and rewards worth cashing in?

Now that we’ve mapped our users and our competition, we have a solid base to define our challenges and plan how to overcome them.

Personas

How Might We

Now that we defined the problems, how do we solve them?

Before opening Figma, I mapped each business goal to the research finding that should drive it, on one page the client signed off on. It stopped every later “what if we…” from reopening a settled decision.

User Flows

I mapped the main journeys so the path from “Onboarding” to “I got paid” stayed as short as possible.

With every decision mapped and signed off, the plan felt solid. Now the real question was how fast we could turn it into something people could actually use.

To build a prototype as fast as possible, we decided to use Google Material for our raw components. We chose Material for a few reasons:

1. Google is the native platform the client already uses, so the look would feel familiar to stakeholders.
2. It let us ship a quick mid-fidelity prototype for review.
3. Material components are built to work with our Angular framework, which saved developer time.

Visual Direction

We pulled colors, typography, and core patterns from the brand’s existing design system to keep Bonus Bowls consistent with its other products.

Key Screens

From Prototype to Shipped Product

I owned the Angular front-end end to end: architecture, component and state design, responsive behaviour, accessibility, and the production deploy, plus the API contract my backend partner and I built against.

It shipped as a real product:

• Bilingual EN / FR-CA: every string, status, and page title flips cleanly.
• Accessibility: WCAG AA, passing AXE checks, with live announcements on receipt status changes.
• A full admin console:the brand’s reviewers approve receipts, manage payouts, and run the product catalog.

The product was built and shipped. But building something and proving it works for real people are two very different things.

Testing happened twice: with real users on the prototype before launch, and with live data after.

Usability Testing

I ran moderated usability tests with 5 participants (3 current users, 2 interested). Each completed a few core tasks:

• Upload a receipt for a recent vet purchase
• Check your balance and start a cash-out

Goal: see how easily people moved through the flow, gather feedback on look and feel, and find points of confusion.

Key Findings

Participants moved through the upload process and described the redemption as trustworthy and fast, mostly because each step confirmed what had happened and the e-transfer payout was something they were already familiar with. The friction showed up in a few places: entering the invoice total before tax, needing a payment method before cash-out, and finding the right product in the catalog.

Changes Made After Testing

The most useful part was watching where people hesitated. The fixes that came out of it:

Final Product

Validating in Production

Testing didn’t stop at launch. I instrumented the live product so the brand could see whether real behaviour matched the research, and to lay a solid base for UX measurement so we can track how well future updates perform.

The GA4 layer tracks the full member journey across 30+ events, with four real conversions: account created, receipt submitted, payout requested, and payment method added. Three decisions kept the data honest:

• Privacy-first: nothing is tracked until the user consents.
• Members vs. browsers: every session is tagged registered or guest, so reports show real members, not all traffic.
• Bilingual in the data: language is captured as a property, so English and French behaviour is visible, not guessed.

A Looker Studio dashboard turns the raw events into plain-language answers that I shared with all stakeholders. Every research number that justified a decision is now a live metric.

Bonus Bowls launched in July 2026, so this data is just starting to flow. The signals below are a first read, not a verdict.

Early signals

Between usability testing before launch and live analytics after, the product could finally answer its own hardest question: is it actually working?

Main Insights

The research was clear: pet parents want cash, won’t tolerate a slow or unreliable receipt process, and need to trust a new program before sharing banking details. Every major decision , cash over points, one-tap upload, security on every money screen, e-transfer payouts , traces back to one of those findings. Designing from data meant every review became “show me the survey answer that says why.”

What I Learned

Thanks for reading. Happy to talk through any part of the process, or just say hi.

carelmaeda@gmail.com