How to write a UX case study when you do not have impressive metrics

Start with the decision you helped make

A portfolio case study should help someone understand how you work. Begin with the user problem, the product context, your role, and the most consequential decision. “I designed a dashboard” names an output. “I helped support managers find urgent cases before changing shifts” gives the work a purpose. Distinguish the team’s work from your individual contribution and state the project’s status.

Use the evidence you actually have

You may have research notes, a support theme, a task walkthrough, usability observations, a prototype comparison, or stakeholder feedback. Each can support a specific claim. None automatically demonstrates a business outcome. If a product never shipped, say that. If you do not have post-launch analytics, describe the evaluation you completed and the result you would measure next. A candid limitation is more credible than a large unexplained percentage.

Show one meaningful before-and-after decision

Select an example that reveals a tradeoff. Perhaps a shorter form removed useful qualification information, so you kept two necessary fields and deferred the rest. Show the old state, the alternative, the reason you chose it, and what remained uncertain. Use captions that explain decisions rather than merely label screens. If the example is a self-initiated redesign, identify it as independent practice, not commissioned client work.

Use a compact structure

Open with context and role. Explain the problem and evidence. Show alternatives and the choice. Describe how you evaluated it. Close with outcome, limitations, and next steps. You do not need a separate page for every method you have learned. Include a journey map or persona only when it affected a decision. Readers should be able to skim the story and then inspect the supporting detail.

Write a defensible outcome statement

Illustrative wording: “In a formative test with four relevant participants, three found the revised status label without prompting. The remaining participant confused archived and completed cases. We changed the label and planned another test. We did not measure post-launch retention.” This is a fictional example, not a client result. It shows the level of specificity that helps a reviewer understand what was learned and what was not.

Protect the work and the people

Use only artifacts you are permitted to share. Remove personal information and follow client or employer confidentiality requirements. Reconstruct a sanitized example when appropriate and label it accurately. If AI helps draft the narrative, supply approved facts and ask it to flag missing evidence. Do not let it invent research participants, quotes, stakeholder approvals, or performance metrics to complete the story.

Try a 30-minute case-study edit

Highlight every claim of impact. Under each, note the source and what the claim actually demonstrates. Replace unsupported claims with a precise observation or an explicit unknown. Then ask a peer to explain your role and the decision after a quick read. For focused feedback on one journey and its design rationale, explore the UX Decision Review; portfolio hiring outcomes are not promised.

Keep exploring

A better-looking redesign is a hypothesis. Here is how to test it.

How to prioritize UX friction without giving your product a made-up score