An AI shopping tool that knows the store as well as the list.

Project
Aisle Assist
Focus
Mobile app design, list management, in-store guidance
What I did
Research, IA, visual design
Context
2023 • UI/UX case study
Single Work Image
Problem


Shoppers are dealing with information overload and disorganization, a pile of lists, loyalty apps, and websites that turns a quick grocery run into an hour of mental overhead. The opportunity was to use AI not as a feature, but as the thing that holds it all together.

Discover

Research

I ran three research threads: a feature audit of four competing apps, a survey of 14 grocery app users, and interviews with regular shoppers. The same unmet needs came up across all three.

Feature analysis

Existing apps handle search and product images well. None connect store-aware ordering and in-trip guidance into a single flow.

  • Store-aware ordering — list reorders to match the aisle layout of whichever store the user picks
  • In-trip guidance — aisle number, stock status, and alternates visible on every product page
Surveys

A survey of 14 list users showed discoverability and list management were the top priorities, far ahead of personalization. That shaped what got cut from v1.

87%
Prioritized product search and filtering, making discoverability the dominant design bet.
64%
Wanted better shopping list management, confirming the list itself was the core design surface.
64%
Valued product descriptions and images, meaning content quality had to match feature quality.
57%
Wanted personalized recommendations, the lowest priority, which informed what to cut from scope.
Interviews

Three themes emerged across conversations with regular grocery shoppers.

“Some stores have their items in different aisles than others. Clothing steamers at Target are in a different section than at Ace Hardware.”

On store variability

“I spend so much time scouring the aisles.”

On time on the floor

“I’m not able to input the quantity by weight.”

On precise inputs

Key Insights

01

Every store is different

A category sits in aisle 4 at one store and aisle 11 at another. A list that doesn't adapt to the store quickly feels useless.

02

Time on the floor adds up

Shoppers said they spend too long searching for items. The design opportunity was to reduce that friction before they're already standing in the wrong aisle.

03

In the moment, less is more

People check the app tired, in a hurry, one-handed. Scannable rows and quiet color do more work than any extra feature could.

What this drove

Made every card lead with the punchline. For each fact, the question became: what's the version of this that sounds like a conspiracy theory but is completely true?

Design

Prototypes

Three screens, each covering a distinct moment of friction: arriving at a new store, routing through it efficiently, and getting help mid-trip without losing your place.

Reflection


This project is from 2023, before AI UX had much of a shared vocabulary. I designed the happy path and caught the gaps late. Today I'd start with the failure cases and lean on existing AI UX frameworks rather than designing from first principles.