Timeline: 4 Months | Sep - Dec 2018

Roles: UX Designer, User Researcher

Methods: User Interviews, Personas, Competitor Analysis, Prototyping, Influence Diagramming

Objective

To create a prototype of a mobile health technology to improve the unhealthy eating habits of the working class citizen by leveraging behavioral change techniques and principles of psychology to influence behavior change.

Problem

Almost half of all American adults have a chronic illness that is related to a poor quality diet [1]. While the benefits to healthy eating are well documented, working professionals who work over 40 hours a week are at increased risk of neglecting their health because of the time burden associated [2]. Intervention strategies that target the working professional need to accommodate these time constraints and the perceived barriers to healthy eating [2]. In addition to diet, working professionals often purchase food made away from home which also contributes to poorer diet quality [3].

Process

Ideation

Targeting the problem for working professionals we began brainstorming possible solutions to target some subset of the challenges working professionals face when it comes to their health. We began by exploring solutions to food waste behavior, eating out, and junk food purchases. Initially we conceptualized a smart fridge application that could address nutrition imbalance by showcasing the nutritional breakdown of food purchases, send reminders when food was close to expiring, and assist users in list making for future grocery purchases. These would leverage behavioral change techniques such as goal setting, self monitoring, and intention formation. This initial solution however and our preliminary assumptions about our target audience needed validation.

Interviews

Generating our initial interview protocols, we conducted 5 semi structured interviews with full time workers across various industries including medicine, technology, accounting, and others. These interviews were critical for understanding our target populations behaviors and attitudes towards grocery shopping and meal preparation. Analyzing our interview notes we discovered a number of key findings that informed our chosen interventions to influence user behavior and give direction to our design. 3 major findings were highly influential in changing our initial design considerations, those major findings were the following:

I was very surprised by users disfavor towards smart fridges. While some conceptually thought of a smart fridge as an interesting idea, they all held serious doubts about the price, inherent value, and maintenance requirements that a smart fridge would carry. Furthermore, the consistency in uninformed decision making across shopping, cooking, and eating behaviors underscored what would need to be the fundamental intervention for our design. Many users habitually go to the store with no real clue of what they will buy, open their fridge with no real idea of what they will make, and eat meals with no real concept of what is healthy. While some users are active about planning their behaviors, most are passive when it comes to shopping and cooking, making decisions in real time which makes it easier for users to make impulse decisions.

Diagramming

Using our research as a guide we mapped out our influence diagrams to showcase the behaviors and contingencies that we aimed to influence. These diagrams highlighted where interventions could be applied and what contingencies they would be affecting.

Recognizing that user behavior had variation due to preferences, attitudes, and context we began to structure behavior change interventions to incorporate these variations dynamically. In this way, the prototype could influence users in different ways based on usage and service multiple user types across contexts. Recognizing users behavioral patterns, attitudes, and variations however with the necessity to narrow the scope to an ever expanding problem, we re-evaluated our initial design considerations and conceptualized new design options to implement in our solution.

Prototype Iterations

With our new design direction, we developed a prototype of a mobile health technology that supports users in purchasing healthier food options by sending recommendations in real time while shopping at the store, and assist users in cooking healthier meals by recommending recipes based on purchase history and estimated inventory.

The design consists of two integral high level functions. Exploring recipe ideas, and generating shopping lists. These two high level functions inform one another to ensure their mutual success. As a user I am able to explore recipes, rate them, and receive instructions on how to make them. If I want to try a particular recipe but don't have the ingredients, I can add that recipe to a shopping list. In my shopping list I can add specific items that I search for, or add items from my purchase history. Once I’m at the store, I can search for the geolocation of items in the store on my list via augmented reality using the camera functionality of my phone, and add reminders to items on my list to notify me once I’m physically close to the item in the store.

These main design functions are mapped to the following behavior change techniques:

• Goal setting >> List making, Recipe selection, Initial on-boarding

• Intention Formation >> List making, Recipe selection

• Barrier Identification >> Reminders, AR location finding, Recommendations

Challenges & Takeaways:

The challenge to this design was identifying how to implement interventions in a way that would address users needs by integrating into their unique behavioral patterns. In particular, the difference in list making behavior was difficult to address because the system would have to compete against more typical and frankly easier tools such as pen and paper, while creating value for non-list makers to secure usage. Additionally, trying to identify what information would be of most value to support buying a new ingredient or cooking a new recipe was challenging because users have ingrained patterns of behavior. However the takeaway I learned is to develop solutions that reduce the burden of a choice by making commitment to a new behavior gradual.

While the main success of the design is to first and foremost get users to make purchases, the follow through is to get users to successfully cook with those items. Moving forward the system needs more testing with real users and more robust changes to influence cooking behavior because the behavioral expense to recipe making is much higher than making a purchase. The system as it currently stands would likely be less successful in influencing the behavior of users who have less formalized cooking experience.

Attribution: Ideation by Daniel Falk from NounProject.com, Interview by Delwar Hossain from NounProject.com, Influencer by Adrien Coquet from NounProject.com, Wireframe by mikicon from NounProject.com, Gauge by Icons Bazaar from NounProject.com, Smart Fridge by Vectors Point from NounProject.com, Planing by Alzam from NounProject.com, Ask by thirddesign from NounProject.com

More Projects