Eat Me

SURVEY DATA

Our team used a mix of quantitative and qualitative user research. We first started by sending out a survey to 56 participants, with a relative distribution of males and females ranging from 18 - 35 years old. 

The pie chart illustrates the most important factors that go into users choosing a recipe. The majority chose to have photos along with the recipe directions, to avoid ingredients they don’t like, and most of them didn’t want a social media aspect for a recipe mobile app either.

EMPATHY MAP

The data we gathered allowed us to generate information to help us inform the user empathy so we can better understand what our users are feeling throughout the cooking process and whether they experience any pain points when searching for a new recipe so synthesizing data into an Empathy Map was a natural next step in our research.

USER PERSONA

Once we were able to get a better understanding of what our users were facing in their cooking and meal decision process, we utilized the data from our empathy map to create a user persona so we generate clarity in who our users were: what lifestyle do they have? What kind of apps do they already use when finding recipes? What are their goals and ambitions as well as frustrations?

USER INSIGHT

A busy professional who is frustrated by the format and content of current recipe sites needs to be able to quickly find recipes using the ingredients and tools they have on hand. They want to save money by cooking instead of ordering take out to make exciting new meals that fit their diet, lifestyle preferences, and cooking skill level.

PROBLEM STATEMENT

Busy young professionals find themselves with a kitchen stocked with an assortment of ingredients, often with little direction as to how to assemble them. They want to cook quick, tasty and novel meals with what is on hand. 

Their current method for finding recipes, primarily through search engines, results in a variety of recipes with varying degrees of instructional fidelity and format. These instructions are often either too detailed or not detailed enough. 

These results either don’t meet the preferences of the users or lead to mistakes during the execution of the meal. How might we create an app that presents recipes in a format that allows them to utilize the ingredients on hand and eases the frustrations the user has with how they currently  source recipes in a fun and useful way? 

ideation

With all of information and insight obtained from our research, we got to the miro board. These are our early ideas centered around the features and must-haves for our app based on the feedback we received. We felt the best approach to figuring out the specific features were by using the “I Like, I Wish, What If” method which resulted in putting this data into a Feature Prioritization Matrix

value proposition

Figuring out what to prioritized from this data allowed us to see the precise value proposition for this mobile app.

Eat Me is a quality recipe app that generates curated recipes for meals that busy professionals can access to discover new meals, share with peers and save for future needs in order to create the ultimate home-cooking experience.

JOURNEY MAP

Now that we had sufficient data organized with these methods, we were able to better understand that the user flow will be when they go through all stages of making a meal: from initial start to finding the recipe and the execution of that recipe in order to complete a meal. We used this journey map to better visualize what the user experiences throughout these stages:

storyboard

Establishing the user journey allowed us to imagine a scenario that our users would typically find themselves based on the data and wanted to use Jeff from our proto-persona to inform what he will experience when using the Eat Me app.

user flow

Now that we were able to define the lifestyle, behaviors, pain points and the gains from our general users, we were then able to create a user flow to understand the pathways the users will take when using our apps and the features we included from our feature prioritization matrix.

competitive analysis

We took a moment to do a competitive analysis because we wanted to understand what other apps are currently available to users that provide a solution to similar user problem and wanted to compare and contrast each of them so that we can understand the areas for improvement, areas that users respond positively, and areas where there can still be innovation within the recipe generation space.

prototyping

Now that we gathered enough research about who our users are, what kind of features we wanted to include in our app, what the user journey will look like, and learning more insight about competitors we could now start the prototype process. Each group member created their own hand sketches that we created lo-fi prototypes with using InVision.

sketches

Using insights from our research, we each sketched individual paper prototypes based off our user flow. 

lo-fi wireframes

Once we compiled the best features from each member’s paper sketches, we were able to come up with a low fidelity wireframe mockup of our Eat Me app.

homepage A/B testing

We found that there were two key directions we could go within the homepage design that can showcase what the users are looking for when they reach the homepage and searching for new recipes with the Eat Me app. So we conducted A/B testing and concluded that the “B” version was the most positively engaged amongst our users.

hi-fi prototype

After our user testing with the lo-fi prototype, we were able to iterate on specific wireframes to enhance the user experience and we were able to translate that into a high fidelity prototype using Figma.

testing and iteration

Objective ​
A) Can users successfully complete onboarding, including building their My Kitchen Profile?
B) Are users able to successfully search for a recipe with ingredients and filter their searches?
C) Determine which version of the app executes these steps more successfully in A/B Testing

Tasks ​
1)  Sign up for account using Email
2) Build My Kitchen Profile
3) Search for a recipe with ingredients you have on hand and filter search as necessary. 

4) Select top suggested result and save recipe.

Feedback
✓ - Testers were successfully able to sign up using Email.
✓ - Testers were successfully able to build My Kitchen Profile, though there were questions regarding options provided.
✓ - Testers confused by initial coaching screens but were able to search for ingredients with little issue​​, though finding filters button was an issue for a few testers.

✓ - Testers were successfully able to select top suggested result and save recipe. Few had layout and format suggestions for recipe page.

X - Tester Failed to get past the loading screen

usability testing matrix and analysis

We conducted usability testing of our hi-fi prototype and wanted to find insights related to which features of the current app design were the most valuable to our users and for our goals as designers.

conclusion

This was an exciting project that allowed our group to explore each step of the UX process in an in-depth manner. We learned many valuable lessons from both our successes and mistakes.

Further research is crucial
We realized how important user research is during this process. For every ideation, for every iteration, we kept referring to our notes and transcripts for the feedback we received during our interviews. For the 6 people we interviewed, the 56 people we surveyed, and the 7 usability tests we conducted, there was so much insight we gathered. We never imagined how many times we would reference any of it, and with more user research we will be able to iterate even more successfully. 

Coding is needed for a better prototype

We learned that the very nature of prototype testing resulted in a number of failures and slowdowns as the user navigated through the app. The loading page confused users and had them slow down or give up all together. The preloaded filter settings and personalization/onboarding confused some testers who were trying to input their own preferences. To simulate true personalization and get better results from testing is to actually code so the app can be more similar to what the user would actually experience.