Stretch — improving an AI grocery app when the data wasn't ready

The vision was bigger than the foundation underneath it. So we made the product better with what it already had.

Role: UX design and product strategy
Timeline: March to April 2026
Team: Product strategist, UX and product strategist
Outcome: Incremental UX improvements; parted on good terms

The problem

Stretch is an AI-powered grocery app that helps shoppers find the best prices on their grocery lists. It already existed. Our job was to clean up the experience and bring in new ideas that made it better for the people using it.

The vision was bigger than the foundation underneath it. Many of the features we imagined depended on data the product didn't yet have.

What we found

User research was limited on this project, so we leaned on product analysis and strategy to guide our decisions. Two things stood out.

Data was the hardest part

Grocery prices change constantly. Getting accurate, current pricing at scale was far harder than we expected, and every feature that promised "the best price" depended on it.

Small improvements took time to land

We proposed modest changes to make the product easier to use, and not all of them were adopted right away. That told us as much about the team's readiness as about the product.

Every feature that promised the best price depended on data the product didn't have yet.

What we changed

Since Stretch already existed, we focused on refining it rather than reinventing it. We cleaned up the experience and made the product clearer and easier to use. We put changes that didn't depend on missing data first.

We started in Figma, then tried moving the work into Figma Make. It was an early experiment in changing how we design.

We also proposed a blog campaign to position Stretch as a thought leader in smart grocery shopping. Strong content could build credibility and strengthen the product, not just promote it.

What happened

The client's data limitations meant they weren't ready for the more ambitious work we were proposing. We helped them make incremental changes that made the experience noticeably better, then stepped away on good terms.

What we learned

A product is only as strong as the data behind it. Ambitious AI features need a reliable foundation first, and it's worth testing that assumption early.
Incremental improvements are still real wins. When a client isn't ready for the full vision, small changes can still make a meaningful difference.
Content strategy can strengthen a product. Positioning a company as a thought leader builds trust and clarifies what the product stands for.