Wholeshare aims to reduce food wastage for co-located individuals by creating shopping groups where individuals can purchase food in bulk for their group and then split food based on personal need. It automatically creates a group shopping list from everyone’s personal lists, notifies the group when someone starts shopping and calculates how much money is owed to the shopper.
I worked closely with four other students and took lead on designing the data collection and user testing tasks, and also contributed to other aspects of the project.
See the core workflows and InVision prototype at the end of the case study.
Duration: 14 weeks
Type: Classwork project for the Personal Health Informatics class.
Target Audience: Small co-located groups such as college students, working professionals, a neighborhood of families etc.
Deliverable: High-Fidelity InVision Prototype and Project Poster
My Role: UX Researcher
Survey Design and Deployment
Brainstorming - Ideation and Pro-Con Analysis
Low Fidelity Sketches
Paper Prototype Testing
High Fidelity Prototype Design
Tools: Mural, Figma, InVision, Paper Prototyping
Design a solution to reduce food wastage for individuals in co-located communities.
We carried out the project in three stages, as shown below. It was important for us to identify and validate our problem space during the research phase, therefore, we conducted a literature survey of academic papers and official food wastage reports. We then distributed an online survey to our participants to record their food wastage trends. We analyzed the survey data to come up with potential solutions and tested them with participants to iterate and refine our design. Finally, we created the core application screens and a high-fidelity prototype.
We started research with a literature survey to identify a problem space where we could design an effective solution for reducing individual food wastage and then deployed online surveys to learn more about the food consumption trends of our target population.
We reviewed academic papers, EPA reports and internet articles to gather information about recent food wastage statistics and existing food waste management solutions. We found that the US wasted more than 96 billion pounds of food annually and nearly ten percent of this food was wasted by individuals. Additionally, few food waste management solutions targeted individual food wastage, and these solutions were either highly localized or placed a high burden on individuals, thereby greatly limiting their access and effectiveness in reducing food wastage.
The lack of effective solutions for personal food waste management was an important discovery that validated our problem space - there was an opportunity to design a solution to individual food wastage that avoid the pitfalls of existing solutions and effectively tackled the problem.
We used two online surveys to capture the food wastage trends of our participants. The first survey was a recurring survey which tracked food purchase and wastage over the course of 6 weeks. The second survey was a one time onboarding survey which captured participant preference regarding sharing or receiving food, deciding when to discard food and preferred method of food disposal. The surveys were distributed to 20 graduate students at the University of Maryland because this demographic was a close approximation of our target audience.
We collected both qualitative and quantitative data in the surveys and used keyword frequency analysis and survey analytics to identify insights from our data. These were as follows
Participants purchased food once every one or two weeks
Participants purchased food in bulk due to cheaper prices but were often unable to finish it before it went bad
Wasted food mostly consisted of short term perishables and was thrown out weekly or bi-weekly
Wasted food was mostly thrown out because it had either gone bad or, the taste was not satisfactory
Participants were willing to share or receive food from friends but not strangers in their community
The ideation phase consisted of brainstorming, prototyping and testing. During brainstorming, we came up with two ideas and evaluated them against our survey insights to choose the one which suited our design goal. We then prototyped our idea by creating low-fidelity sketches and user scenarios. Finally, we moved on to testing where we asked participants to complete core application tasks using a paper prototype.
Based on our research findings (reduce individual food wastage, reduce wastage due to excess food) we came up with two preliminary ideas
Idea #1 - Connect individuals with excess food to individuals who want food.
Idea #2 - Make it easier for a group of individuals to shop in bulk and then divide the food amongst themselves.
To identify which idea best suited our project goals, we conducted a pro-con analysis based on our survey insights. The analysis indicated that the second idea - facilitating bulk shopping for a group - was a better choice because it was directly utilized the existing behavior of bulk shopping, reduced cost and excess food by allowing people to split food based on need and created awareness of personal food needs. Furthermore, due to the largely self-regulated nature of the solution, it did not require significant infrastructure and was possible across a range of socio-economic situations.
We continued brainstorming for design ideas which would aid the shopping experience and identified the most popular ideas by voting on them. The most relevant design ideas were as follows
Allow users to create a personal food list which tells the group what items they require
Maintain a group food list which is a combination of food items from each user’s individual food list
Let one person shop for the group, and notify the rest of the group when they started and completed shopping
Make splitting group bills easier by adding receipt scanning and comparing it against personal food lists
During the prototyping stage we created user scenarios to envision how people would use Wholeshare to fulfill their food needs during the day. These scenarios led to the identification of core application tasks, which created an information architecture. We visualized the architecture using low-fidelity sketches and after a round of iteration created paper prototypes for user testing.
The user scenarios helped us understand how users would use the app to satisfy their food shopping needs, shop for themselves and the group, and accommodate quick shopping trips for critical items. Based on our user scenarios, we created low-fidelity sketches which visualized core application tasks such as adding food to a personal list, viewing the group shopping list with a breakdown of food per person and using a checklist while shopping. We organized these sketches into an information architecture which formed the foundation of our application.
We tested the information architecture and user interaction flow with five participants using paper prototypes. We asked the participants to complete core tasks within the app with little to no help.
Based on the observations and feedback received from testing, we made the following changes to our design:
Users did not understand the function of the urgent checkbox since it was not well explained by the application UI. We removed this option to avoid confusion.
Users could guess but were not absolutely certain about the function of the “private” checkbox. We made this more explicit, by making it switch between “private” and “shared” states to indicate when their items were shared with group list.
We added a “+” icon to the group list for users to indicate they also wanted that item. This achieved the same function as adding the food item to their personal list but as a faster one-click action.
The “Collect” screen was universally confusing because the name did not represent its function. We renamed it to “Purchased Items” and added a badge to indicate when a payment was due.
Finally, we designed the application screens in Figma and created an interactive prototype using InVision. A video showcasing the most common workflows is at the end of the section.
Navigating this project from the broad idea of food waste reduction to its final form as Wholeshare proved to be a robust exercise in both user research and social health informatics. My key takeaways from this project were the importance of good survey design and user testing. Both of them were critical in understanding our users and challenged the key assumptions we made during the design process.
While our work for the class project has concluded, we recognize that there is still more we can do in this project. Most importantly, our high fidelity prototype remains untested. Testing it and iterating on the design would be the first step.
Additionally, in its current state Wholeshare offers little incentive for members of the group to shop for others. It relies on peer encouragement to ensure that students in a group will shop for each other. We imagine the group, ideally consisting of friends, will devise a system which has everyone take turns to shop. However, this is still an assumption and may not occur. Our initial ideas to incentivize shopping for others include small monetary adjustments in favor of the shopper and badges which display the number of times someone has shopped in the group. Finding the right balance between these elements should incentivize shopping to the extent that most groups will find Wholeshare to be a sustainable solution for their bulk shopping needs.