The UX audit checklist for 2026, with the source behind every item
A sourced UX audit checklist in six sections: Nielsen's 10 heuristics with three to five evaluators and a severity scale, WCAG 2.2 AA including the new criteria, Core Web Vitals at the 75th percentile, forms and checkout, content and navigation, and the legal dates to know.

Key takeaways
- 01A UX audit checks a product against usability heuristics, accessibility standards, performance thresholds and real user data, then ranks problems by severity.
- 02NN/g recommends three to five independent evaluators for a heuristic evaluation; in Nielsen's early research, single evaluators found only 35% of the usability problems.
- 03Test against WCAG 2.2 AA, which added nine success criteria and removed 4.1.1 Parsing; WCAG 2.2 is also the international standard ISO/IEC 40500:2025.
- 04Measure Core Web Vitals at the 75th percentile of real visits: LCP 2.5 s or less, INP 200 ms or less, CLS 0.1 or less.
- 05Know the dates: the European Accessibility Act applied from 28 June 2025, and the US ADA Title II web rule now applies from 26 April 2027 or 26 April 2028 (checked 23 October 2026).
A UX audit checks a site or app against usability heuristics, accessibility standards, performance thresholds and real user data, then ranks the problems by severity. Use three to five independent evaluators, test against WCAG 2.2 AA and Core Web Vitals at the 75th percentile, and turn findings into a prioritised backlog.
Most UX audit checklists online are lists of good intentions: "check navigation", "make it accessible", "test on mobile". They rarely name a standard, an evaluator count or a threshold, so two teams using the same list get different results. This checklist links every line to the source behind it: Nielsen Norman Group for heuristics and severity, W3C for accessibility, web.dev for performance and Baymard for checkout. Legal dates were checked on 23 October 2026. Accessibility law is summarised here as context, not legal advice.
What should you prepare before a UX audit?
An audit without goals produces a long list of opinions. Set these up first:
- Business goals and success metrics. What should the product do better: sign-ups, completed checkouts, fewer support tickets, faster task completion?
- Key user journeys. Pick three to six journeys that matter most, such as "find a product and buy it" or "invite a teammate". The audit walks these, not every page.
- Users and contexts. Who uses the product, on which devices, and in what conditions? A field app used on a phone in sunlight needs different checks from a desktop dashboard.
- Analytics and field data access. Analytics funnels, search terms, error logs, support tickets and Core Web Vitals field data (from the Chrome UX Report or your own monitoring).
- Scope and standards. Agree in writing which pages, platforms and standards are in scope (for example, WCAG 2.2 AA and Core Web Vitals) and who receives the findings.
- A severity scale. Agree how findings will be rated before anyone starts, so priorities are not argued afterwards.
How do you run a heuristic evaluation with Nielsen's 10 heuristics?
Heuristic evaluation is the core of most UX audits. Jakob Nielsen developed the heuristics "in collaboration with Rolf Molich in 1990", and NN/g still publishes them as the standard set:
| # | Heuristic | What to check |
|---|---|---|
| 1 | Visibility of system status | Loading states, progress, confirmation after actions |
| 2 | Match between the system and the real world | Plain language, familiar terms, logical order |
| 3 | User control and freedom | Undo, cancel, back, easy exits from flows |
| 4 | Consistency and standards | Same words and patterns for the same things; platform conventions |
| 5 | Error prevention | Constraints, sensible defaults, confirmation before destructive actions |
| 6 | Recognition rather than recall | Visible options, remembered choices, helpful defaults |
| 7 | Flexibility and efficiency of use | Shortcuts, saved items, bulk actions for frequent users |
| 8 | Aesthetic and minimalist design | No competing elements; content that supports the task |
| 9 | Help users recognize, diagnose, and recover from errors | Plain-language error messages that say what to do next |
| 10 | Help and documentation | Help that is easy to find, task-based and short |
How many evaluators? NN/g's guide to conducting a heuristic evaluation says "three to five people should independently evaluate" the design. The reason is in Nielsen's original research: "single evaluators found only 35 percent of the usability problems". Different people spot different problems, and pooling them gives far better coverage.
The process, step by step:
- Brief evaluators on users, goals and the journeys in scope.
- Each evaluator works alone, walking each journey at least twice: once to learn the flow, once to inspect each screen against the heuristics.
- Each records every problem with the screen, the heuristic it breaks, evidence (a screenshot) and a suggested fix.
- Merge findings into one list and remove duplicates.
- Rate each problem for severity, ideally independently, then agree a final rating.
The severity scale. NN/g's severity ratings combine three factors: frequency, impact and persistence. The scale runs from 0 to 4:
| Rating | Meaning (NN/g) |
|---|---|
| 0 | Not a usability problem at all |
| 1 | Cosmetic problem only: fix if extra time is available |
| 2 | Minor usability problem: low priority |
| 3 | Major usability problem: important to fix, high priority |
| 4 | Usability catastrophe: imperative to fix before release |
Which accessibility checks belong in a UX audit (WCAG 2.2 AA)?
Accessibility is part of usability, so a UX audit should include it. Use WCAG 2.2 at level AA as the target. W3C's What's new in WCAG 2.2 says it became a W3C Recommendation on 5 October 2023 (the current edition of WCAG 2.2 is dated 12 December 2024), and on 21 October 2025 W3C announced that WCAG 2.2 "is now an approved International Organization for Standardization (ISO) standard: ISO/IEC 40500:2025" (W3C news).
WCAG 2.2 adds nine success criteria:
| Criterion | Level | What to check |
|---|---|---|
| 2.4.11 Focus Not Obscured (Minimum) | AA | Sticky headers, cookie banners and chat widgets don't hide the focused element |
| 2.4.12 Focus Not Obscured (Enhanced) | AAA | No part of the focused element is hidden |
| 2.4.13 Focus Appearance | AAA | Focus indicators are large and high-contrast |
| 2.5.7 Dragging Movements | AA | Anything done by dragging also works with a single tap or click |
| 2.5.8 Target Size (Minimum) | AA | Tap targets are big enough or spaced apart |
| 3.2.6 Consistent Help | A | Help options appear in the same place across pages |
| 3.3.7 Redundant Entry | A | Users aren't asked to retype information they already gave in the same process |
| 3.3.8 Accessible Authentication (Minimum) | AA | Sign-in doesn't depend on memorising or transcribing; paste and password managers work |
| 3.3.9 Accessible Authentication (Enhanced) | AAA | Stricter version of 3.3.8 |
W3C also notes that "4.1.1 Parsing is obsolete and removed from WCAG 2.2", so don't spend audit time on HTML validation for that criterion.
Beyond the new criteria, check the basics that still fail most often: text and non-text contrast, alt text, form labels and error messages, heading structure, keyboard access to everything, visible focus, captions for video, zoom and reflow to 320 CSS pixels, and screen reader announcements for dynamic content. Combine automated tools (which catch only part of the problems) with manual keyboard and screen reader testing. For a full conformance audit, see website accessibility.
Which performance checks matter (Core Web Vitals)?
Slow or jumpy pages are a usability problem. Use Google's Web Vitals thresholds, measured at the 75th percentile of real page loads, split by mobile and desktop:
| Metric | Measures | "Good" threshold |
|---|---|---|
| LCP (Largest Contentful Paint) | Loading: when the main content appears | 2.5 seconds or less |
| INP (Interaction to Next Paint) | Responsiveness: delay after taps, clicks and key presses | 200 milliseconds or less |
| CLS (Cumulative Layout Shift) | Visual stability: unexpected movement | 0.1 or less |
Checks to include:
- Use field data first (Chrome UX Report or real user monitoring), then lab tools to diagnose causes.
- Check the key journeys, not just the home page. Product, search and checkout pages often perform worst.
- Look for the usual causes: oversized hero images, render-blocking scripts, third-party tags, web fonts, heavy JavaScript on interaction, and images or ads without reserved space.
- Retest on a mid-range Android phone on a mobile connection, not only on a fast laptop.
What should you check in forms and checkout?
Forms are where users give up. Baymard's checkout usability research puts the average documented cart abandonment rate at 70.19% on its research page; its cart abandonment list gives 70.22% as of 22 September 2025, averaged across 50 studies. Not all of that is fixable, but much of the friction is.
Checks to include:
- Field count: remove every field you don't need. Combine name fields where possible.
- Labels and errors: visible labels (not placeholder-only), inline validation after the user leaves a field, and error messages that say how to fix the problem.
- Input types: the right mobile keyboard for email, phone and numbers; autocomplete attributes for addresses and cards.
- Guest checkout: don't force account creation before purchase.
- Costs up front: shipping, taxes and fees visible before the final step.
- Redundant entry: "billing same as shipping" by default (WCAG 3.3.7).
- Authentication: allow paste and password managers; offer passwordless or email codes where suitable (WCAG 3.3.8).
- Trust and help: clear returns information, security cues near payment, and help in the same place on every step (WCAG 3.2.6).
How do you audit content, navigation and information architecture?
- Labels: navigation uses the words users use, confirmed with search logs and support tickets, not internal jargon.
- Findability: key tasks are reachable from the home page and from search; check zero-result searches.
- Hierarchy: one clear H1 per page, logical headings, and the most important content first.
- Wayfinding: breadcrumbs or clear section markers on deep pages; the current location is always visible.
- Readability: short sentences, plain language, scannable lists, and link text that makes sense out of context.
- Consistency: the same component and the same words for the same action everywhere.
- Empty, loading and error states: every list, search and dashboard has a helpful message when there is nothing to show or something fails.
Which legal dates should a UX audit flag?
A UX audit is not a legal review, but it should flag accessibility deadlines that affect the product. These are the dates as published on official sites, checked on 23 October 2026:
| Law or standard | Date | What it means |
|---|---|---|
| European Accessibility Act (EU) | Applied from 28 June 2025 | Accessibility requirements for many consumer products and services, including ecommerce, in the EU (AccessibleEU) |
| ADA Title II web rule (US state and local governments) | 26 April 2027 for entities serving 50,000 or more people; 26 April 2028 for smaller entities and special district governments | WCAG 2.1 AA for web content and mobile apps; dates extended by an interim final rule published 20 April 2026 (ADA.gov) |
| ISO/IEC 40500:2025 | Approved October 2025 | WCAG 2.2 as an international standard (W3C) |
WCAG 2.2 builds on 2.1, so an audit to 2.2 AA also covers the 2.1 AA criteria, apart from 4.1.1 Parsing, which 2.2 removed. Ask your legal adviser which obligations apply to you; an audit tells you where the product stands, not whether you are legally compliant.
What does the full UX audit checklist look like?
| Area | Check | How to test | Standard or source | Default severity if failed |
|---|---|---|---|---|
| Heuristics | 3–5 evaluators review key journeys independently | Individual walkthroughs, merged list | NN/g | n/a |
| Heuristics | System status is visible after every action | Walk each journey | Nielsen heuristic 1 | 2–3 |
| Heuristics | Users can undo, cancel and go back | Try to exit each flow mid-way | Nielsen heuristic 3 | 3 |
| Heuristics | Error messages explain the fix | Trigger each validation error | Nielsen heuristic 9 | 3 |
| Accessibility | Focused element never hidden by sticky UI | Keyboard through each page | WCAG 2.4.11 (AA) | 3 |
| Accessibility | Drag actions have a single-pointer alternative | Try without dragging | WCAG 2.5.7 (AA) | 3 |
| Accessibility | Tap targets meet minimum size or spacing | Measure on mobile | WCAG 2.5.8 (AA) | 2–3 |
| Accessibility | No retyping of information already given | Complete each multi-step form | WCAG 3.3.7 (A) | 3 |
| Accessibility | Sign-in works with paste and password managers | Test login and sign-up | WCAG 3.3.8 (AA) | 3–4 |
| Accessibility | Help appears in a consistent place | Compare pages | WCAG 3.2.6 (A) | 2 |
| Accessibility | Contrast, labels, keyboard and screen reader basics | Automated scan plus manual testing | WCAG 2.2 AA | 3–4 |
| Performance | LCP 2.5 s or less at p75 | Field data, mobile and desktop | web.dev Web Vitals | 3 |
| Performance | INP 200 ms or less at p75 | Field data, then lab traces | web.dev Web Vitals | 3 |
| Performance | CLS 0.1 or less at p75 | Field data, then lab traces | web.dev Web Vitals | 2–3 |
| Forms | Guest checkout and costs shown early | Complete a purchase | Baymard | 3–4 |
| Forms | Only necessary fields; visible labels | Count and review fields | Baymard; WCAG 3.3.2 | 2–3 |
| Content | Navigation labels match user language | Search logs, card sort or tree test | NN/g heuristic 2 | 2 |
| Content | Zero-result searches reviewed | Site search report | Analytics | 2 |
| Legal | Applicable accessibility dates flagged | Scope review with your adviser | EAA; ADA Title II | n/a |
Severities are starting points. Adjust them using NN/g's frequency, impact and persistence factors for your product.
UX audit vs usability testing: what's the difference?
| UX audit (expert review) | Usability testing | |
|---|---|---|
| Who finds problems | Experienced evaluators, against heuristics and standards | Real or representative users doing real tasks |
| What it finds | Breaches of known principles, accessibility and performance failures | Where actual users get confused, stuck or give up, and why |
| Speed | Days | Usually longer: recruiting, sessions, analysis |
| Best for | Fast, broad coverage before a redesign or test | Validating priorities and fixes with real behaviour |
They work best together. Run the audit first to fix obvious problems cheaply, then test the riskiest journeys with users. If the audit uncovers conversion problems, a CRO programme can test the fixes with real traffic.
How can Elgrith help?
Elgrith has delivered 1,000+ projects for 250+ clients since 2015, from our New Delhi headquarters and client centers in New York, London and Shanghai. Our UX audit team runs heuristic evaluations with multiple independent evaluators, WCAG 2.2 checks and Core Web Vitals reviews, and gives you a severity-ranked backlog. For full accessibility conformance work, see website accessibility; to fix what the audit finds, our UI/UX design team takes it from there. Accessibility findings are audit and remediation guidance, not legal advice. You get a fixed written quote after a free call. Book a call or start a project.
FAQ
Frequently asked questions
Sources 12 linked
- 01Nielsen Norman Group — 10 Usability Heuristics for User Interface Designnngroup.com(opens in a new tab)
- 02Nielsen Norman Group — How to Conduct a Heuristic Evaluation (25 Jun 2023)nngroup.com(opens in a new tab)
- 03Nielsen Norman Group — Theory of heuristic evaluations (1 Nov 1994)nngroup.com(opens in a new tab)
- 04Nielsen Norman Group — Severity ratings for usability problemsnngroup.com(opens in a new tab)
- 05W3C WAI — What's new in WCAG 2.2w3.org(opens in a new tab)
- 06W3C — Web Content Accessibility Guidelines (WCAG) 2.2w3.org(opens in a new tab)
- 07W3C WAI news — WCAG 2.2 approved as ISO/IEC 40500:2025 (21 Oct 2025)w3.org(opens in a new tab)
- 08web.dev — Web Vitals (updated 31 Oct 2024)web.dev(opens in a new tab)
- 09Baymard Institute — Checkout usability researchbaymard.com(opens in a new tab)
- 10Baymard Institute — Cart abandonment rate statistics (updated 22 Sep 2025)baymard.com(opens in a new tab)
- 11AccessibleEU — The EAA comes into effect in June 2025 (31 Jan 2025)accessible-eu-centre.ec.europa.eu(opens in a new tab)
- 12ADA.gov — Fact sheet: Title II web and mobile app accessibility rule (read 23 Oct 2026)ada.gov(opens in a new tab)

