Levi's Assortify

Reimagining internal tools

Intro/the brief

Levi’s the global jeans brand, started working with Red Badger at the beginning of 2021. Levi’s employees (fashion merchants in this case) are heavily reliant on internal cumbersome tools to get their clothing lines to market. The main goal from the business's perspective was to take the current go-to-market processes of 14 months down to just 9 months with a new all-in-one internal digital solution called ‘Assortify’ (Spotify for jeans).

The team

The cross-functional team consisted of a tech lead, a project lead, software engineers, a UI designer, and two Senior UX designers (myself included). With additional tech support and a project manager from Levi’s side. As the 2nd Senior UX Designer on the team, I was responsible for taking requirements and translating them into fully working and tested features.


  • Empower Levi's to work more efficiently through Agile ways of working
  • Get users using Assortify for key parts of their clothing delivery cycle
  • Develop a centralised tool to manage the whole Assorting process


  • Levi’s HQ is based in San Francisco, which meant we had a two-hour window of working time with the client each day
  • Levi’s operates on a waterfall model, we would need to educate the client in Agile and highlight the best ways of working
  • Merchants (our main users) have built up strong mental models making it hard to introduce new ways of doing things

User problem(s)

  • The clothing go to market process is long and complicated
  • Users juggle between many different types of software in their day to day
  • Documents set up by different teams are convoluted and difficult to edit

The process

Utilising the Double Diamond design model with a heavy emphasis on the Discovery quadrant, we opted for a fully Agile way of working to release updates in two-week sprints and testing/learning from an MVP approach. Features and changes would be planned in either a 6 month roadmap or from common themes in user comments and research.

The team and I spent countless hours with the client refining our discovery process, this served as an education piece and ensured we maximised our output. Once the Discovery process was in place and working well, we worked on a Miro template to reuse for other client projects.

Requirements gathering

Requirements gathering was a point of friction from early on, a mix of complicated subject matter and uneducated stakeholders had us sharpening our consultancy skills. The team and I experimented with different processes, often the client would write super complex requirements instead of one-liner user problems. We introduced items such as:

  • A context former: a document the client could fill out in their own time and would cycle through a list of questions
  • Brain walking exercises: each team member would write down their interpretation of the problem
  • User problem breakdown: as a team go through a user problem and present what we know back to the client

Opting for the Brain Walking exercise, as this was the most successful approach for the team, we eventually had a deeper shared team understanding of each new problem and where that particular problem came from (business vs user).

User research

The project started with multiple rounds of subject matter expert interviews, gaining insights into the merchants day to day processes, and pain points, building a cohesive understanding of the clothing go-to-market processes. All of the interviews were documented in Dovetail, which allowed us to tag the transcripts, and create themes in the participant’s comments.

Taking our learnings from the user interviews, the team and I developed a large service blueprint, this helped the team (and the client) understand the user's ways of working at a glance.

From there, we were able to create detailed personas to provide context when designing, and for new joiners to gain a deeper understanding of our users.

A large portion of research and feedback was gathered asynchronously. Adding questions to a board in Miro worked best. We would often set stakeholders (and users) homework to do in our absence, we would leave comments and questions on wireframes in Miro.

To combat the time differences we tested wireframe concepts and Figma clickable prototypes with European merchants, as well as conducted multiple rounds of guerrilla testing with Red Badger employees. Levi’s would also hold their internal UAT testing with users, this consisted of the Product Manager taking the users through the new features and getting verbal feedback. A Microsoft Teams channel was set up so that users had a space to voice their concerns, give feedback and give suggestions.

Research insights

  • Users day to day consists of switching between many different types of software
  • Master files are created by other teams but often abandoned because they’re difficult to use
  • A huge amount of time was spent manually updating Excel spreadsheets, PowerPoint documents, and presenting groups of products using Visulon
  • Most of the software the users are using doesn’t integrate, which leads to a lot of copying and pasting of important information

Building an MVP

Once the research was gathered we had a good basis to start creating an MVP of Assortify. Taking inspiration from the commerce website we built a rough design library of components. The MVP allowed us to rapidly test iterations of interactions and experiences, however, after senior stakeholders weighed in, it was evident that we would need to completely re-style the MVP. We hired additional UI design resources to revamp the Design Library and to give the site an overhaul.

UX Heuristics evaluation

Cracks started to appear within Assortify, as the site grew bigger and bigger with more features. We quickly realised that we needed to re-design some of the interactions to ensure the site was still easy to use and made sense to the users. I decided a heuristics evaluation was the best course of action, this allowed us to once again educate the client on UX best practices and stress test our current site.

In a couple of days I was able to:

  • Run the entire product through the 12 UX heuristics
  • Present back the most urgent findings and showcase how they might fit into the roadmap
  • Alleviate bias and personal opinions to focus on the best UX practices


More times than not, we would go straight to wireframing in Miro instead of sketching with pen and paper. We came up with a suite of wireframe templates in Miro to rapidly build designs.

There were many benefits to choosing this method:

  1. It had a big impact on productivity, speeding up time to get the ideas down
  2. The wireframes matched the live site
  3. The UI designer could keep wireframes separate from the Figma design library clean and tidy
  4. Showcasing the ideas to the stakeholders was easy and they could leave feedback directly on the board


Handovers consisted of taking the UI designer and the tech lead through a story breakdown. Telling a story, going from the user problem through to client-approved wireframes ensured our design thinking was communicated effectively. At this stage in the process, the team and I would uncover outstanding questions, decide what we could build in one sprint, what would need to be parked for version two of the feature, and then involve the client for one last check creating the finalised UI.

The outcome

In just a month we had a fully functional MVP with a structured process for consisting improving Assortify. The product quickly grew with more and more helpful features tested and making a positive impact on users day to day. The team and I were able to identify when we outgrew the MVP structure and quickly pivot to adjust usability. Red Badger gave Levi's insight into Agile methodologies and ways of working, we were able to rapidly produce and test features within a traditionally waterfall environment. Assortify caught the attention of senior stakeholders, impressed with the rate at which we were implementing features, and nominated the Assortify team for an internal award.

Feedback from users was extremely positive and excited with the new features and changes we were making. Here are a few stand-out features that delivered the most impact to users:

  • Search for assortments and products
  • Data injection from other data product software
  • A mood board page whereby users could organise key looks and present assortments (groups of products)
  • Different view options allowed the users to find key product information at the click of a button
  • Made assortments collaborative, private, and or public

Final takeaways

The team and I made a positive impact through educating the business on Agile methodologies, shaking up the current ways of working by constantly releasing features at pace, listening to users and giving them a voice to leave feedback/suggestions. We were able to alleviate some of the stresses users faced getting their clothing lines to market, which now takes only takes around 10 months instead of the original 14.

After a year on the project I really got to sharpen my consultancy skills and make a dramatic impact to Levi’s digital strategies and ways of working.

Key learnings

  • Experiment with all sessions in your toolkit until something works for you and the client
  • Working within a two hour time window per day is tough, but there are practical ways to get things done
  • Your have to be very efficient with you time and gather feedback/research asynchronously
  • Being diligent with the structure of Miro boards, making them super clear and easy to read

What I'd do differently

  • Stop and reflect on our achievements, we shipped an incredible amount of features in a year but it wasn’t celebrated as much as it should have been
  • Don’t rush into a project, get a feel for how the project is going by talking with different members of the team
  • Ensure we had a project manager/business analyst within a similar timezone
  • Work out the best ways of working with the client early on, and nail down all the processes you’re both happy with
Next project →

More projects...


Discovering what’s possible

Nando’s ID

Reworking login & sign up flows

Nando’s hackathon

A mini game whilst you wait