Skip to content
PromotEdge International

Home Guides UX audit checklist

Guides · Template

UX audit checklist: what to check before a redesign

A redesign that starts without an audit usually changes how a site looks and keeps how it fails. A UX audit finds what stops people completing their tasks, with evidence, so the redesign brief lists problems to solve. This checklist covers eight areas, from goals and data to content, with the standards behind each. It is part of our guide to the UX design process.

Reviewed by Anindita BarikUpdated 12 Nov 20264 min read

The short answer

What should a UX audit check?

A UX audit checks whether people can do what they came to do, and why not. It starts with the business goals and the data (analytics, support tickets, search terms), reviews the screens against Nielsen's 10 usability heuristics, walks through the key tasks step by step, checks accessibility against WCAG 2.2, tests on a phone, measures speed against Core Web Vitals, and reviews content and labels. The output is a ranked list of problems with evidence, which decides what a redesign must fix.

Key takeaways
  • Audit before you redesign, so the redesign fixes known problems instead of moving them.
  • Start with goals and data; then review screens against Nielsen's 10 usability heuristics.
  • Walk the three to five key tasks step by step, and watch about five real users try them.
  • Check accessibility against WCAG 2.2 Level AA and speed against Core Web Vitals.
  • Rank findings by severity and how many users they affect, with evidence for each.

Need this done for your business? A person reads every message and replies.

Ask for a call back

UX audit checklist at a glance

The eight areas, what each checks and the yardstick behind it. Our own working checklist, with sources named.

Our own working checklist. Yardsticks: NN/g usability heuristics and five-user testing; W3C WCAG 2.2; web.dev Web Vitals; Pew Research Center (Oct 2026).
AreaWhat to checkYardstick
Goals and dataMeasures, key events, drop-offs, searchesGoogle Analytics key events
HeuristicsEach key screenNielsen's 10 usability heuristics
Key tasksThree to five tasks, step by stepAbout five users per test round
NavigationLabels, wayfinding, searchOur own working practice
FormsFields, errors, confirmationOur own working practice
AccessibilityContrast, keyboard, labels, targetsWCAG 2.2 Level AA
Mobile and speedTasks on a phone; LCP, INP, CLSWeb Vitals: 2.5 s, 200 ms, 0.1
ContentHeadings, answers, button textOur own working practice

Check eight areas: goals and data, usability heuristics, key tasks, navigation and structure, forms, accessibility, mobile and speed, and content. Back each finding with evidence (analytics, a recording, a test session, a standard) and rank them by severity and reach. The ranked list becomes the redesign brief.

1. Goals and data

Before looking at screens, gather:

  • The business goals for the site or product, and the measures that track them.
  • Key events counted correctly in analytics; Google Analytics lets you mark events as key events.
  • Where users drop off in the main funnels.
  • What users search for on the site, and what they ask support.

2. Usability heuristics

Review each key screen against Jakob Nielsen's 10 usability heuristics (1994, last reviewed January 2024):

  • Visibility of system status: does the screen show what is happening?
  • Match with the real world: does it use the user's words, not internal terms?
  • User control and freedom: can people undo and leave?
  • Consistency and standards: do the same things look and work the same way?
  • Error prevention: are mistakes hard to make?
  • Recognition rather than recall: are options visible on screen?
  • Flexibility and efficiency: are there shortcuts for frequent users?
  • Aesthetic and minimalist design: is anything on screen that does not help?
  • Help users recover from errors: do error messages say what went wrong and how to fix it?
  • Help and documentation: is help there when needed?

3. Key tasks

List the three to five tasks that matter most, such as finding a product, requesting a quote or completing setup, and walk through each step by step, noting every point of friction. Then watch real users try the same tasks; Nielsen's article on testing with five users (2000) argues that about five per round find most problems.

4. Navigation and structure

Check:

  • Menu labels use words users recognize.
  • Users can tell where they are and how to get back.
  • Important pages are no more than a few clicks from the start.
  • Search works and handles misspellings.

5. Forms

Forms are where tasks end. Check:

  • Only the fields you use; labels above fields.
  • Errors shown next to the field, with how to fix them; no lost input.
  • The right keyboard on phones for each field.
  • A clear confirmation of what happens next.

6. Accessibility

Check key screens against WCAG 2.2 Level AA, the current W3C Recommendation (US rules such as ADA Title II cite WCAG 2.1 AA, which 2.2 includes):

  • Text contrast and readable sizes.
  • Every function works by keyboard, with visible focus.
  • Images have text alternatives; form fields have labels.
  • Targets meet WCAG 2.2's minimum of 24 by 24 CSS pixels, or its exceptions.

7. Mobile and speed

Pew Research Center's mobile fact sheet (19 October 2026) finds 91% of US adults own a smartphone. Check against web.dev's Web Vitals:

  • Every key task works on a real phone, one-handed.
  • Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint within 200 milliseconds and Cumulative Layout Shift within 0.1.
  • No pop-ups that cover the page on small screens.

8. Content

Check:

  • Headings say what each page or section is for.
  • Text answers the user's questions in the order they ask them.
  • Buttons say what they do.
  • Out-of-date or duplicate pages are found and listed.

Ranking the findings

Rate each finding by severity (does it block the task, slow it or annoy?), by how many users it affects, and by the effort to fix. Attach the evidence: a screenshot, an analytics figure, a test clip or the standard it breaks. Fix blockers on the current site now; the rest becomes the redesign brief.

Conclusion

A UX audit checks goals and data, usability heuristics, key tasks, navigation, forms, accessibility, mobile and speed, and content, then ranks the findings with evidence. Run it before any redesign so the new design fixes known problems, and use Nielsen's heuristics, five-user tests, WCAG 2.2 and Core Web Vitals as the yardsticks.

PromotEdge International is an international strategy, technology, marketing and design company, and our delivery team is in India. Our UI/UX design team runs this audit and designs and tests the fixes, and our website redesign work rebuilds from the findings.

Facts we rely on: see our company facts for what we state about ourselves and where each fact comes from.

Common questions

  • A structured review of a website or product to find what stops people completing their tasks. It looks at goals and data, usability heuristics, key tasks, navigation, forms, accessibility, mobile use, speed and content, and ends with a ranked list of problems, each with evidence, that guides fixes or a redesign.

  • Eight areas: goals and analytics data, Nielsen's 10 usability heuristics, a step-by-step walk through the key tasks, navigation and structure, forms, accessibility against WCAG 2.2, mobile use and Core Web Vitals, and content. Each finding needs evidence and a severity rating.

  • Yes. Without one, a redesign changes how a site looks and often keeps how it fails. The audit gives the redesign a brief made of known problems with evidence, and lets you fix blockers on the current site while the new design is under way.

  • For a typical marketing site or one product area, in our own working practice, a few days to two weeks, depending on how many key tasks, screens and user types are in scope and whether user testing is included. Larger products are audited flow by flow.

  • Visibility of system status; match with the real world; user control and freedom; consistency and standards; error prevention; recognition rather than recall; flexibility and efficiency; aesthetic and minimalist design; helping users recover from errors; and help and documentation. Jakob Nielsen published them in 1994.

  • Yes. Our UI/UX design team runs this audit, ranks the findings with evidence, then designs and tests the fixes, and our website redesign work can rebuild from them. We quote after seeing the site or product, and publish no rate card.

Brands we have worked with

  • Fine Organics
  • Weichai
  • Novozymes
  • Dozco
  • SANY
  • Prezent
  • Honest AI
  • SCENNE
  • Sara International
  • Vega MPS
  • Desi Machine Pvt. Ltd.

Work delivered from India. Results are shared one to one, under NDA.

See all clients

Talk to us

Tell us what you need

We use these details only to reply to you.

Phone or WhatsApp (optional)
Who is it for? (optional)

Rather talk now?

How we work

  1. 01

    Tell us the goal

    A rough note through the form on what you sell and what needs to change.

  2. 02

    We study your market

    How people find you today, where inquiries get lost, then a call in your working day.

  3. 03

    A written plan

    The work, who does it and how it is measured. Nothing starts until you approve it.

  4. 04

    The work starts

    Our team in India does the work, with calls set in your working hours.

  5. 05

    Reports and visits

    Every number we report shows its source and date range, and we travel to our clients every 3 to 4 months.

Chat with us