Aroma Case Study: IOS/Android App
Project Outline
Role: Product Designer
Responsibilities: User Testing, UI Design, UX Design
Company: Aroma.io
Team: 3 Engineers, 1 Marketer, 1 Designer
Brief
Aroma is a food-focused search engine that was created to solve the age-old question “What restaurant sells ‘X’?” During my time running Aroma I, with the help of my team crafted the user interface and experience for the iOS and Android native application. I used human-centered design to discover problems, interview users, design, test and iterate thoughtfully.
The Problem
We found through user interviews that people have a hard time finding a place that serves the food they want to eat.
After doing 15 user interviews we found each user had a different method to finding a joint, from using delivery service apps like Postmates/Uber Eats/Waitr to facebook groups and asking for help on Twitter.
15 / 15 Users found their method very hit & miss.
Competitors
Google: Searching for food is problematic due to the broadness of the search.
Yelp: Very outdated results and the restaurant’s name is only searchable.
Zomato: Best of the three but their search was limited to restaurant name and cuisine.
Our Solution
Create a better way to search and find food by allowing for dish names and ingredients to be searchable. We accomplished this by creating a database and filling it with dozens of menus; each item dynamic and searchable.
Metrics for Success
- Decrease the time in which it takes a user to find a place to eat (15 to 30 Minutes)
- Increase the rate of finding a restaurant that has the food they crave (70% of interviewers say they don’t often find the place)
The Results
These results were found after talking to the original 15 plus testing another group of 15.
Positive
• 24/30 Found a restaurant serving what they were craving.
• 20/30 It took less than 10 minutes to find the place they wanted
Negative
• The search is clunky and at first confusing.
• The catalog was small so searches outside of the main city were hit and miss.
Iteration
Moving forward we created a more human-friendly search. We did this by allowing non-keywords to be used without affecting the results to make the search feel more natural. We also added predictive keyword suggestions to help with spelling and errors. Last we made the search auto populate so it searched while you typed for ease of use.
We tested the prototype for this iteration and the user feedback was very positive.
• 0/15 Needed prompts for help
• 15/15 Found this prototype an overall huge improvement
Sun-setting
A month after the launch of Aroma two of the developers and the marketer from the team received offers from other companies that they just couldn’t turn down. With a lack of developers and manpower, we deiced to sunset Aroma.
What I learned
As a leader and a designer, I made dozens of mistakes. The result of these mistakes has truly humbled me. I have learned a lot but they all culminate into what I call now my three pillars of design.
One: Obsess over the problem
I would often forget the main problem I am solving and try to solve symptoms of the root problem. This created some major bloat in Aroma that really set the timeline back.
Two: Test as often as possible
My understanding of testing wasn’t very strong and I believed I needed a nearly fleshed out product before I could test it. This resulted in learning things later rather then soon and pushing the timeline back.
Three: Ego does nothing
Too often I felt the need to defend my work and be right in everything I do. Being right or wrong doesn’t make me a better designer, and worrying about it doesn’t make me a good teammate either.
Moving forward I will obsess over the root problem, fearlessly test often, and seek truth in my assumptions without pride.