15 minutes, report the moment they finish
UX writer skills assessment
Twelve written situations, about fifteen minutes, that measure whether a UX writer fixes the interface or just the sentence: error states, destructive confirmations and copy covering for a broken flow.
The gap
What you can’t see when you hire a ux writer
A writing sample shows the final sentence. It never shows what the screen looked like before, or what they refused to paper over.
The case starts here
The Support Lead: “The error message is confusing. Can you make it friendlier?”
Every case opens with a request that already contains its own answer, because that’s how requests arrive at work. What we measure starts with whether they notice.
What this case shows you
- Whether they rewrite the copy or flag the broken flow behind it
- How they handle a destructive action
- What they do with a legal constraint they can't remove
- Whether they write for the reader or for the brand
- How they decide what not to say
The kind of call they make
Situations a ux writer runs into
Twelve short situations like this one, drawn from the day-to-day work of the role. The live situations are confidential, because an assessment you can rehearse stops measuring anything.
Example situation
Support tells you an error message is confusing and asks you to make it friendlier. Looking at it, the error fires when a user tries to save a form that has timed out, and their work is lost.
What you get
A recommendation for your ux writer role
You mark what this particular job can’t do without. The report opens with a clear call for that job, shows where the candidate stands on each capability you chose, and gives you three questions to ask in the interview.
Marta R. · UX Writer
Leads with both of the things you named
Reaches first for writes for the person in the moment and last for calls the same thing the same name, and works at a different pace from your team.
Why these matter for a ux writer
- Spots when the brief is wrong
- For this role the brief is usually 'make this sentence better' when the sentence is covering for a broken flow. Noticing that, and saying it without refusing the work, is the job.
- Sweats states and edge cases
- UX writing lives almost entirely in states nobody designed carefully: errors, empties, confirmations, and the dialogue in front of a destructive action. That last one is where careless copy costs real data.
- Explains decisions outside the team
- Mostly to legal, support and brand, all of whom have a claim on the words. A writer who can't negotiate with them writes whatever the last email said.
Against what you already do
What each format sees, and what it misses
| Format | What it shows you | What it misses |
|---|---|---|
| Writing samples | Voice, economy, and whether they can hold a tone. For a writing role this is solid evidence and the fastest signal available. | What the screen looked like before. A sample is the finished sentence; it can't tell you whether the interface needed a better sentence or a different flow, or which of those they argued for. |
| Editing exercise | Craft at sentence level: cutting, clarifying, and fixing a tone that’s off. | Whether the sentence was the problem. Handed a confusing error message, the valuable answer is often 'this error shouldn't be reachable', and an editing exercise has no way to accept that answer. |
| Portfolio walkthrough | How they explain their choices and whether they can defend a cut. | A comparable answer. Every candidate brings a different project and gets different follow-ups, so at the end there’s nothing to line up side by side. |
| This assessment | Whether they fix the interface or just the sentence, across error states, destructive confirmations and legal constraints they can't remove. Twelve identical situations, read against your role. | Prose quality. Keep the writing sample; it answers a question this doesn't. |
How it works
Three steps,
with no calls and no setup
What must they be good at?
- Spots when the brief is wrong
- Knows what the data supports
- Moves without being told
- Sweats states and edge cases
One credit per candidate, spent only when they start. The first is free (see what the packs cost), and nothing expires, so a role on hold costs nothing.
Questions
Assessing UX writers, in practice
No. A sample shows prose quality; this shows whether they fix the interface or just the sentence. Both matter for this role and the sensible process asks for both.
No drafting from scratch. Comparing free-form copy across six candidates is the subjective judgement this is meant to replace, and their drafting ability is already visible in the sample.
About fifteen minutes, on any device, whenever suits them. The report reaches you the moment they finish.
They can try, and it helps them very little. There's nothing to write or polish, and a model can't know what this particular person would do. Read past the headline.
No. You name what the role can't do without first, and the report is read against it. A writer assessed for a heavily regulated product and for a consumer app get different recommendations from one sitting.
One credit per candidate, spent only when they start. The first is free with no card, and credits don't expire.
Related reading
Read these before you decide
8 min
Product manager interview questions worth asking
A question discriminates when a strong and a weak candidate give visibly different answers to it. Most standard product manager questions fail that test because the good answer is public and can be rehearsed. The ones that work put the candidate in a situation with no correct option (a request that contains its own answer, a stakeholder who won't move, a deadline that halves) and read what they reach for first.
Read it8 min
The PM case study interview: what it measures and what it misses
A product case interview measures structured thinking, communication under mild pressure and comfort with ambiguity. It doesn't measure what someone refuses, what they kill, how they behave when a plan breaks, or how they hold a position with an executive. Run as a take-home, it also ranks how much unpaid time a candidate had available that week.
Read it10 min
How to tell if a design portfolio was generated with AI
You usually can't tell from the artefact itself, and detection tools don't work reliably. What you can do is look for missing specifics (real constraints, abandoned directions, disagreements, numbers that are inconveniently precise) and then ask one interview question that only someone who was in the room could answer.
Read it