UX writer domain knowledge interview questions

As asked

What reading level do you target for consumer product copy, and does that change for an enterprise B2B product? What frameworks or tools do you use?

Sample answer outline

Candidates should cite Grade 6 to 8 as the commonly accepted consumer target based on US literacy statistics, note that enterprise does not automatically mean higher complexity (experts still benefit from clear language), name tools like Hemingway Editor, Flesch-Kincaid, or Gunning Fog Index, and acknowledge that reading level is one signal among several.

Expect these follow-ups

• What does reading level not capture that you still need to care about?
• How do you apply reading level guidance to a product with very technical subject matter?

As asked

Name three UX writing patterns you reach for regularly, explain what each is, and give a situation where each is appropriate and one where it is the wrong choice.

Sample answer outline

Strong answers cite patterns like: progressive disclosure (right for reducing cognitive load, wrong when users need full information to make a decision), front-loading (right for scanning, wrong when context must precede the instruction), or the confirm-and-continue pattern for destructive actions. Candidates should show depth of understanding not just ability to name patterns.

Expect these follow-ups

• What pattern do you see misapplied most often?
• Where do you look when you encounter a situation you have not written for before?

As asked

How does cognitive load theory inform the choices you make about how much copy to include on a screen?

Sample answer outline

Candidates should explain intrinsic, extraneous, and germane load without using jargon, translate these to practical writing decisions (chunk information, remove redundant copy, use parallel structure to reduce parsing effort), and show that less copy is not always better if it leaves users without essential context.

Expect these follow-ups

• When is adding more copy the right decision for reducing cognitive load?
• How does cognitive load theory apply to error messages?

As asked

How do you define the boundary between UX writing and content strategy, and what happens at the overlap?

Sample answer outline

Candidates should articulate UX writing as the execution layer (the actual strings) and content strategy as the systems layer (what content exists, for whom, maintained how, measured how). Strong answers show they think at both layers and describe how the two inform each other, especially in terms of channel strategy and content lifecycle.

Expect these follow-ups

• Have you ever had a project where the content strategy work changed what you wrote?
• What skills does a UX writer need to grow into content strategy work?

As asked

What is a mental model in the context of UX writing, and can you give an example of copy you changed to match the user's mental model rather than the product's internal model?

Sample answer outline

Candidates should define mental models as the user's existing understanding of how something works, explain the tension between internal product naming (which reflects how the system works) and user vocabulary (which reflects how the task works), and give a concrete example such as renaming an "Account" to "Profile" or a "Journey" to a "Trip" based on how users talk.

Expect these follow-ups

• How do you find out what users call something if you have no direct research access?
• What do you do when the product name is the mental-model mismatch and you cannot change the name?

As asked

When do you use "your" versus "my" in UI copy, and how do you decide? Give an example of each and explain the principle.

Sample answer outline

Candidates should explain: "your" is the application speaking to the user (conversational, human), "my" implies the interface represents the user's perspective (navigational labels like "My Account"), the convention that navigation tends to use first person while instructional copy uses second, and the importance of consistency within a surface.

Expect these follow-ups

• What happens when a screen mixes both and a user reads them in sequence?
• How do you handle this in languages where the distinction does not map neatly?

As asked

When does a visual UI element need an ARIA label written by you rather than just relying on its visible text, and how do you write a good one?

Sample answer outline

Candidates should identify cases where visible text is absent or insufficient: icon buttons with no text label, images that are controls, close buttons labeled only with an X, tables with ambiguous headers, grouped inputs. Strong answers explain that ARIA labels should describe function or result, not appearance, and name the WCAG criterion (4.1.2 Name Role Value).

Expect these follow-ups

• How do you review ARIA label quality in a product when you cannot read code?
• What is the most common ARIA labeling mistake you see in products you have audited?

As asked

What is information scent, and how do you use it when writing navigation labels and link text?

Sample answer outline

Candidates should define information scent as the cues that tell users whether following a link or entering a section will lead to what they need, explain how specific labels outperform clever or brand-driven ones, describe how link text should describe the destination rather than the action, and mention how breadcrumbs and section headers reinforce scent.

Expect these follow-ups

• How do you test whether a navigation label has good information scent?
• Give me an example of a low-scent label you have improved.

As asked

How does the typographic environment affect how you write, and what is your process when a type style makes your copy harder to read?

Sample answer outline

Candidates should discuss line length and its effect on reading speed, how all-caps affects readability and should influence whether you use it for UI elements, how font size affects the tolerable word count on a screen, and how to advocate for typographic changes when the design is working against the copy rather than adjusting words to work around a bad design.

Expect these follow-ups

• Have you ever advocated for a type change because of a copy readability problem?
• How do you handle a design system where the display style is all-caps for all navigation items?

As asked

What is pseudolocalisation, and how would you use it early in a product development cycle to catch copy and layout problems before real translations exist?

Sample answer outline

Candidates should define pseudolocalisation as replacing source strings with padded, accented equivalents to simulate translation expansion (German averages 30% longer than English), explain that it reveals hard-coded widths and truncation before any real translation costs are incurred, and describe advocating for it as a build step in internationalised products.

Expect these follow-ups

• How do you get engineering to set up pseudolocalisation if it is not in their current workflow?
• What ratio of expansion do you plan for when writing source English strings?

As asked

Walk me through the experience of a screen reader user navigating a page you have written. What does that tell you about how to structure headings, links, and form labels?

Sample answer outline

Candidates should show they have actually used a screen reader (NVDA, JAWS, VoiceOver) and understand how users navigate by heading, landmark, or interactive element rather than reading top-to-bottom. This should translate to: descriptive headings that work out of context, link text that describes the destination (not "click here"), form labels that make sense when read alone, and unique button labels.

Expect these follow-ups

• What is the worst screen-reader experience you have ever tested in a product you worked on?
• How do you convince a team to invest in screen-reader testing before launch?

As asked

What is content governance and what does it look like in a large organisation with 30 product squads all writing their own microcopy?

Sample answer outline

Candidates should describe governance as the system of people, processes, and standards that keep content consistent and on-strategy, explain the tension between centralised control and squad autonomy, and propose lightweight governance mechanisms: shared style guide with enforcement tooling, copy review as part of design sign-off, periodic audits, and clear ownership of the standards document.

Expect these follow-ups

• How do you make governance feel like a resource rather than a bureaucracy?
• What do you do when a squad ships copy that violates the standards without review?

As asked

Some design systems include text as tokens, for example a "label.button.primary" token. What are the advantages and limitations of treating copy as a token, and where do you think this model breaks down?

Sample answer outline

Strong answers appreciate the benefit for consistency in highly standardised UI elements but identify the limit: copy changes with context (a primary button that says Save in a form should say Confirm in a dialog), localisation makes single string tokens insufficient, and the model privileges predictability over usability in edge cases. They should also mention the handoff complexity.

Expect these follow-ups

• How would you advocate for copy flexibility within a component that has a default text token?
• Where do you think design tokens and string IDs in localisation platforms should connect?