Sage - a food sharing app

Time frame:Feb-May 2013

Role:Researcher, Designer, Prototyper


I conceptualized, designed and built a prototype mobile app for buying and selling home cooked meals, during Harvard Grad school.

The goal of the project was to design a solution to address the problem of unhealthy eating habits of students by providing an easy, affordable alternative.

I led all aspects of the product design including user research, solution conceptualization, idea validation, wireframing, usability testing and prototyping. After project completion, I presented the business idea and the accompanying product as a start-up pitch to the VCs which was very well received. Also, I presented the project during an exhibition at the Harvard School of engineer and sciences. We received great feedback on the simplicity and functionality of the app.


Sage - Case Study


Objective

Unhealthy food habits represented a significant struggle for large number of students at Harvard. The situation was exacerbated with students’ busy academic and social lives, and lack of readily available healthy meal options. The project’s goal was to develop a solution to address this need.


Design process

User Research on Eating Patterns

We conducted user research to identify the user’s needs and pain points. We used three research techniques: (1) User Photographs, (2) Survey on Mechanical Turk, and (3) Social Media research

Camera study - we recruited 6 students to capture photographs of their eating habits for two days. We asked the recruited students to click pictures of food procurement decisions (bur or not), food consumption decisions (eat or not), people they interacted with during procurement and consumption and other interesting food related instances.



Mechanical Turk - we conducted a need finding survey on mechanical turk and gathered responses from 51 users. The survey questions were intended to gather data on parameters that govern eating habits such as: (1) Cost, (2) Cooking skills, (3) Demographic, (4) Cleanliness, (5) Dietary restrictions and special needs. We also included questions to understand the social behaviour around food habits such as willingness to share food.

Social Media research - finally, we analyzed data on social media such as twitter, yelp, reddit, etc. to get an understanding of the parameters that influence eating habits.

The research provided us with incredible insights and the main findings are listed below:

  1. Cooking is time intensive and people often do not have enough time to invest in cooking for themselves
  2. Cooking requires skill and people often have an inertia on picking-up the skill
  3. People who cook food, claim that they often throw away the leftovers
  4. Healthy eating is expensive, both when purchasing from restaurants or buying raw materials for cooking themselves
  5. People have special food needs which are governed by health restrictions, weight concerns, cultural norms and personal preferences
  6. Clean food is a crucial need for consumers

Ideation of solutions

After the initial research, we delved into the solution ideation phase of the project by leveraging the key findings. We generated several solution ideas and after a pros/cons analysis narrowed down to one solution: Food Sharing.

Conceptually, this meant that people cooking meals for themselves could sell a part of their home-cooked meal to someone else. At the consumer side, this will provide easy, economical and readily available options to get a healthy home cooked meal.


Early Validation

Before we proceeded with the solution implementation, we decided to perform an early user validation to ensure that our solution idea was sound. Specifically, we wanted to test people’s willingness to cook food for others.

We posted an advertisement on Craigslist asking people to provide a home-cooked meal for $10-15 per meal. We received a highly positive response as people responded with enthusiasm and some even included photographs of their previously cooked meals. This gave us good confidence that there was both supply and demand available for this idea.



Product Design

We went through several iterations of design and created prototypes using Sinatra for testing. Some of the examples are below:


Iteration 1


Iteration 2


Iteration 3


Usability Testing



We recruited three people to perform usability testing of the first design prototype. We identified four key tasks for the testing, and the problems discovered through each task are listed below:

Task 1: Registering and signing to the app

  1. New users that registered on the app were required to re-enter login details after registering before they could use the app
  2. No indication of successful registration on the app

Task 2: Choosing a meal and contacting the chef

  1. No ability to sort meal pictures based on popularity, price, food type
  2. Confusion with tab title of ‘Chef’
  3. Ability to select a meal by clicking on a picture. Instead the first version required the user to click on the chef

Task 3: Ordering the food with vegan meal preference

  1. Confusion with the term ‘Diners’. Users were instead reading it as ‘Dinners’
  2. No confirmation for successfully ordering a meal

Task 4: Uploading pictures of the food on the user profile

  1. Lack of clarity on where exactly the picture needs to be uploaded on the profile

Based on the above testing findings, we introduced the following changes in the designs:

  1. Added confirmation notifications for user actions such as Registration completed, Order posted, Photos added, and New profile picture.
  2. Changed terminology to drive clarity, e.g. changed ‘Chefs’ to ‘Providers’ and made ‘Diners’ the home screen
  3. Display of posts would be automatically filtered by relevance criteria such as popularity, proximity to consumer, etc.

Impact
  1. I presented the business idea as a start-up pitch to a group of VCs, which was very well received.
  2. Also, I presented the project during an exhibition at the Harvard School of Engineering and Sciences. We received great feedback on the simplicity and functionality of the app.