Umi is an iOS app that brings you fresh, home-cooked meals from your neighbor’s kitchen right to your dinner table. As the lead UX designer on this project, I collaborated with other designers to create a prototype for the MVP of the application.

Umi hero image

The challenge

Umi approached us to create a prototype of their MVP that was simple, efficient and showcased the home-cooks as well as their food. It also had to allow families to easily order several portions for several days at a time. The design needed to reflect homemade delicious food. Because this was a different type of business model, it had to stand out.

Our solution

After conducting a thorough analysis of the market and understanding the customers, we designed a prototype of the app that included a food menu for the next 6 days, a feed with home-cooks in their kitchens and the ability to add items to the bag and check out. We also focused on the experience of entering a virtual kitchen.

Umi prototype

Analyzing the market

Understanding the food delivery market in New York City is a fairly long and complex task so we settled on 6 popular food applications. After studying some of the direct competitors’ flows, we compared their features and target audiences.

Competitor's Analysis Competitor's Analysis Features

Click to Zoom

Talking to our users

User surveys were sent out to screen potential Umi customers for interviews. Once we had identified our primary users, we talked to them to find out more about their habits when ordering food. We asked them how they planned their meals, what food delivery services they used and what they liked or disliked about them.

Afiinity map

We learned that they will order delivery when:
• The restaurant location is near their homes and/or offices
• They can have it delivered quickly
• Their schedules are busy

And that they typically:
• Order easy, quick and healthy meals
• Order food 1 to 4 times a week
• Use either their phone, computer or sometimes call directly

User stories

Based on these results, we formulated user stories for 2 different types of customers in order to keep the customer in mind and prioritize the features accordingly.

“As a parent with two kids, I want healthy, tasty and quick meals so that everyone is satisfied and we can spend more family time together.”

“As a working professional in NYC, I want a curated meal delivered fast, so that I don’t have to think about what to eat when I get home late.”

Prioritizing the features

By doing this, we knew what to focus on for our Minimum Viable Product. We also spoke with engineers and asked them to sort the features by implementation difficulty. This allowed us to prioritize them in a realistic way.

Feature Prioritization

User flow

We focused on the main user flow: a customer opens Umi to order meals for several days and then goes to check out.

User flow


Starting with a paper prototype, we put together a low-fidelity version to test several feature concepts. It was important for us to prove the concept before diving in too deep. Then, we iterated on it based upon usability testing and users feedback.

Paper Prototype Medium fidelity prototype

Test early and often

After testing each of our designs several times, we discovered that:
• Users didn’t realize until checkout that they could order for the whole week
• They weren’t sure if the $15 price listed on the home feed was for 1 person or for the whole family
We changed the copy to address their issues and updated the final prototype. You can view it here.

What I learned

Timebox everything

Managing our time to create several flows for the MVP proved to be challenging. We wanted to be thorough in our research as well as in our design but we had a limited time to do so. We followed our project plan and timeline by timeboxing our day-to-day tasks to be more efficient.

Finding the right users to interview is essential

We wanted to speak to parents to learn more about their behaviors so we had to be selective in our interview process. We filtered specific types of users to talk to with surveys and personal connections.