Project 2 Report — Blue Apron

By Josh Reinitz

Project Overview

On the surface, project 2 was built on increasing user engagement on Blue Apron’s application by allowing users to post their own recipes and then allowing other users to vote on those recipes. If the recipes received enough votes, the recipes would make it to a featured page and then become available for purchase on Blue Apron. The goals for our team were to screen and interview users and potential users to understand what might help to motivate them to post and contribute to this initiative. From there we needed to develop personas and create a prototype based on the personas goals and needs. Following that, our team entered into two rounds of usability tests, iterations, and a presentation for Blue Apron providing our findings.

My role

UX Designer

Timeframe

10 Days

Limitations, Parameters, Resources, and Materials

Initial limitations on the project were that personas were not defined and as a result our team had to understand not only how to integrate the features that Blue Apron was asking for but, we also needed to understand how to integrate them in ways users would actually use. In order to design our prototype we began by creating a screener/survey that was used in order to narrow our interview sample to the people we thought would provide the best insights for our users. Our screener contained questions pertaining to knowledge of Blue Apron, cooking habits, as well as methods of contribution that the users make online. Our team placed each user response into either a “hot” category, meaning the user is currently active and meets a high level of criteria, “warm”, meaning the user met a mid level of our criteria and may be a former member, and “cold”, meaning the user was not a good interview candidate.

Initial Problem Statement

Our initial assumptions were that users would be motivated to post their recipes online if there was a sort of perceived competition. If user’s knew that they could make it to the featured recipes page, they would want to share their success online with others in their social circle. This led us to the initial problem statement of

“Blue Apron users want to engage more with each other by posting and voting on recipes online.”

“How might we integrate a feature that motivates users to want to create and share their own recipes online?”

How did we confirm or refine our initial assumptions?

Initially, our team recruited five interviewees from our “hot” collection of screeners. These were users who currently use a meal kit delivery service and are likely to contribute online whether on social media or sharing through other sources online. These users quickly showed our team that although the service was a benefit in their lives, it was not enough to keep them there for an extended duration. Many of these users were only on the site due to some form of incentive and as such planned to leave once that incentive was over. However, they also gave us major insights which included

• Users wanted some form of education in the kitchen with their Blue Apron subscription.
• Any recipe from Blue Apron should be easy and low effort.
• Users wanted an easy to navigate application.
• Users wanted recommendations on recipes and cooking techniques.
• Users had a desire to save money by using Blue Apron.

User Interviews

Our user interviews were essential in uncovering what motivates users to create and post online. Also, other considerations such as cooking habits and decisions on food choices were added to help our team understand the behaviors and goals of the users we were interviewing. Using these questions, our team really tried to understand pain points of the user in an effort to figure out how to integrate an effective feature that would be of great value to the user and engage them in the Blue Apron application. some of the questions that were included were:

• Walk me through your last experience using Blue Apron or any other meal delivery kit service.
• Tell me about some things you like to do when cooking.
• How do you decide what to cook?
• Who cooks the most in your household?

Affinity Mapping

Following our first round of interviews our team was armed with a series of insights, we began mapping out the features that our users spoke of and by way of affinity mapping we began to find patterns in the users behaviors, goals, needs, and pain points. By doing this, our team was able to understand trends the users had and create attributes that were associated with those needs, behaviors, goals, and pain points. We focused on three groups in particular that were highlighted not only in our affinity maps, but also in our interviews insights.

Design Studio

Following our affinity mapping session, we moved into a design studio in order to see the insights that we discovered materialize in each team members vision. Each team member sketched for ten minutes, presented their ideas for two minutes, and gave feedback for one minute. This was repeated for three sessions and following this we believed we had an idea on how to build in certain features and needs for the users.

Our team going over design studio iterations.

MoSCoW Maps, Feature Prioritization Matrix, and App Maps

After our teams design studio, we moved into understanding what features were essential to the prototype and were on a low effort vs. high effort scale. First, we began by creating a feature prioritization matrix which allowed us to visualize features on an X and Y axis of essential vs. nice to have and low effort/expense vs. high effort/expense. Once we had these features plotted out on the Feature Prioritization Matrix, we moved them to our MoSCoW map which places each feature into a category of, must have, should have, could have, and won’t have. This insures that the features that are added into the prototype are only those that are critical to the users needs and goals. Finally, in order to understand the user flow within the application, we created an app map which allowed the team to visualize how the user would move through the features of the application.

Mid-Fi Prototype Link: https://invis.io/CKNVJMUA6WZ

Round 1 Usability Tests and Resulting Iterations

Following the creation of our mid fidelity prototype based on insights given to us from our user interviews, our team moved into its first round of usability testing. The users were given the mid fidelity prototype and were asked to complete a series of tests. These tasks were designed to identify if the features that were built into the prototype were effective at providing the a vehicle to post a recipe as well as create a place for users to engage on the Blue Apron platform. The users tasks were as follows:

Task 1: Add a recipe within the Blue Apron application that includes 5 ingredients.

Task 2: Vote on a recipe that includes chicken and tomatoes.

Task 3: View how many likes a recipe that you posted has accumulated.

Following our synthesis of the data from the usability tests we gathered the following insights:

• Users found adding a recipe was easy for them and they enjoyed adding the recipe to the app however, they needed a back button to navigate through certain features.
• Users wanted to be able to edit their submissions after making them.
• Users wanted to be able to not only search for recipes.

Round 2 Usability Tests and Resulting Iterations

Following the previous round of testing and iterations, our team implemented a back button for each page, an edit button on all submissions, and a much more responsive elastic search function. These designs were all added as a result of our insights from round 1. Moving into round 2 we brought the application into a much higher fidelity. By doing so, we noticed the users had a much easier time moving through the application as many design features were much clearer to the user. Also, in the second round of usability testing, we noted that users flowed through the tasks with a much more direct path of entry and exit. Finally, we noted that the users all reported that they enjoyed the process and felt that by performing the tasks they would of liked to seen more of a community built into the application. The key insights we gained from round 2 of usability:

• Users reported a bit of confusion regarding the voting system on the application, as such, a clearer system was created where the icons were enlarged.
• Key features of the recipes page were moved, for instance, the featured recipes page tab was moved up to the very top of the page making it much more visible.
• Finally, in order to create a sense of community, a comment section was added in which users could interact directly with one another.

Hi-Fidelity Application Link: https://invis.io/P7NWKO7KFVY

Reflections

This project was a great exercise in understanding how to navigate feature implementation based on the users needs. Each user test showed us that the user really wanted to be involved with this application and desired to post online provided the method was easy for them. Ultimately, each user validated Blue Aprons request in their own unique way. However, the most fascinating thing about this project was that each user ultimately desired the same thing towards the end of our second round. Each user expressed the desire to be involved in a community while learning new techniques and skills on Blue Aprons site. Said differently, the true goal that was not expressed until the very end was that by creating the community the users desired, Blue Apron would gain the user engagement they were seeking in the initial prompt. Whats even more interesting is that this is not an insight exclusive to this project and has shown up in many of the other projects I have been involved in.