Build-a-Pack Design Sprint


The Problem

Underwear Packs are a core part of the MeUndies business model. Customers who purchase packs have one of the highest lifetime values (LTV) for the business, only coming in second to membership subscribers. However, Packs were underperforming in sales. Our goal for 2018 was to elevate packs to 25% of our total revenue for the year, but we had been stuck well below that for a while. As a result, key stakeholders were looking to shift our strategic focus to improving the overall pack purchasing experience for our customers.

Possible Issues

  • Packs are isolated and difficult to find–unless a customer uses the Packs menu in the primary navigation, they are likely going to miss the pack products all together.

  • People aren’t easily able to compare pack pricing to single unit pricing.

  • People want the latest, most popular color options. The colors that come in a pack are usually aged inventory with units available in a person’s selected style and size.

  • People must select a preview link to see what colors will be shipped in their pack.

  • Packs are often returned because people receive colors they do not like in their pack.

MeUndies’ First Design Sprint

The team at MeUndies was familiar with the Design Sprint model popularized by Google Ventures, but had never run one in-house. People embraced the sprint because it made sense for the project: it was a short, experiment that would produce immediate results. We could test riskier ideas and get qualitative customer feedback within a week.

The Sprint Team

We assembled a team comprised of product experts who would be able to carry the project through following the completion of the sprint. The team included: 

  • Gavin Potts, Product Manager (Decider)

  • Jackie Eaton, UX Designer (Facilitator)

  • June Honkiatkhajorn, Sr. UI Designer

  • Gray Gilmore, Sr. Software Engineer

  • Nadia Hruska, Director of Inventory Planning

  • Jeff Camera, Director of Merchandising

We couldn’t bring everyone into the day-to-day work of the sprint, so the gaps in our knowledge was made up by interviewing our C-Suite Executives and other experts in Customer Experience, Consumer Insights, Brand, Retail, and Operations.

Our Goal, Sprint Questions, and Prototype

Through expert interviews and experience mapping, we were able to identify the following sprint goal:

Create a fun, flexible pack purchasing experience for customers who are not interested in membership.

We also spent some time thinking through ways we could potentially fail in reaching our goal by listing out our sprint questions. We came up with many, but the following questions are a selection of a few that guided our design and prototype at the end of the sprint:

  • What if we don’t have enough color options to include in a custom pack?

  • What if customers don’t understand the value of customizing a pack?

  • What if people can’t figure out how to start a pack and add colors to it?

Our sprint questions resulted in a design that didn’t lock people in to specific pack quantities (e.g., 3-pack). Instead, pack savings were unlocked at specific increments–$19 a pair for 3 units or more, $18 a pair for 7 units or more, and so on. We hoped that this model would give customers some flexibility when it came to filling their pack with colors, because it would allow them to add as many as units as they want to it. Alternatively, if we’d asked customers to decide up front on their pack size, they could get to the color selection step and find that there were not enough colors they like to fill their pack.

Design Sprint Prototype Screens

Design Sprint Prototype Screens

Customer Feedback

By the final day of the design sprint, we had a created a clickable prototype using Sketch App and InVision. We recruited five participants from our customer database–a mix of men and women, from Los Angeles, who were not subscribed to our membership program, and had not made a purchase within the last 6-months.

Overall, the prototype tested well with some minor usability issues that could be addressed in the design following the sprint. Some of the issues identified included:

  • Most participants thought the savings meter at the top of the screen was an interactive element, which they tried to click and drag during testing

  • Many participants thought the peace sign icon, which marks progress the amount savings accrued, was counting the number of items in their pack (e.g., 2 pairs)

  • Many participants missed the dollars saved amount in the bottom left corner of the screen, because they were focused on the total price of their pack

  • Although participants were able to learn how to add items to their pack, the stepper element on the selected swatch caused some initial friction. At first some people didn’t realize that selecting the + icon would add a color to their pack.

Mobile production version of Build-a-Pack

Desktop production version of Build-a-Pack

The Numbers & Results

At the beginning of the sprint our goal was simply to create a fun, flexible pack purchasing experience for customers who are not interested in membership. Outside of the sprint, the business was focused on increasing the number of units per order (UPO) and the % of revenue made up by packs.

The Build-a-Pack MVP launched in early October 2018. Since then, we’ve learned:

  • UPO is up to 2.67% from 2.37% the previous month (compared to 2.54% in October 2017)

  • Build-a-Pack now leads MeUndies pack sales (over set packs and mystery packs)

  • The average order value (AOV) of a Custom Pack is $143

What’s Next

Although Build-a-Pack has proven itself very successful since launch, there is still room for iteration and we are continuing to explore opportunities for improvement. After going live, data was showing that most people who were not completing the flow were dropping off at the color selection step. Further user testing has revealed that this may be a result of not having enough color options to choose from when building a pack. Participants in the study were easily able to complete the flow on desktop and mobile devices, but reported significant disappointment with the limited color options available to them. Many of these participants reported they would not purchase a custom pack for this very reason.



Desktop, Tablet, Mobile Web


  • Expert Interviews

  • Journey Mapping

  • Sketching

  • Prototyping

  • Moderated User Testing


  • Whiteboard

  • Sketch App

  • InVision