A UX audit checklist gives you a repeatable way to review an existing website or app against real usability problems, then rank the fixes by business impact. Work through goals, data, navigation, mobile, checkout and accessibility in that order, score each issue, and you will know what to fix yourself and what to hand to a design agency.
| Audit area | What to check | Evidence to collect | Typical first fix |
|---|---|---|---|
| Goals and scope | One business goal, three to five key journeys | Baseline conversion, bounce and exit rates | Write the goal down before touching the design |
| Navigation | Menu labels, depth, search, orientation, calls to action | Click maps, internal search terms, task tests | Rename vague labels, cut menu depth |
| Mobile | Tap targets of at least 24 by 24 CSS pixels, forms, speed, sticky bars | Real-device walkthroughs, mobile funnel data | Enlarge targets, fix input types |
| Checkout | Guest option, field count, cost surprises, payment methods, errors | Step-by-step drop-off, recordings | Remove forced account creation |
| Accessibility | Text contrast of at least 4.5:1, keyboard use, labels, focus states | Manual keyboard pass, automated scan | Fix contrast and form labels |
| Scoring | Severity 0 to 4, impact against effort | One shared findings sheet | Ship the high-impact, low-effort items first |
Most business owners reach this topic after a symptom shows up. Traffic holds steady, but enquiries drop. Mobile visitors leave on the first screen. Baskets fill and then empty. This guide walks through a practical audit you can run in a few working days, using tools you probably already have.
A UI/UX audit is a structured review of an existing website or app that finds usability problems, rates how serious they are, and lists what to fix first. It is a diagnosis, not a redesign. You need one when conversions fall, users complain, or you are about to brief an agency.
The audit combines three kinds of evidence. Heuristic review means checking screens against established usability principles, such as Nielsen's heuristics. Behavioural data shows what people actually do. Accessibility checks show who your design leaves out.
An audit reviews what exists using expert judgement plus data. A usability test puts real users in front of the product and watches them. A redesign changes the product. Run the audit first, because it tells you whether you need a test, a small fix or a full rebuild.
Pick one business goal for the audit, such as more completed orders, more demo requests or more activated app users. A vague goal like "better UX" produces a vague report. A single goal tells you which screens matter and which can wait.
Then list the three to five journeys that drive that goal. For a shop it might be landing page to product page to basket to payment. For a service business it might be home page to service page to contact form.
Write down the numbers you will judge success against. Without a baseline you cannot prove an improvement later. Capture these for each journey:
Looking at screens first is the most common audit mistake. You end up arguing about colours while the real leak is a broken form field. Collect evidence from several sources, then let the evidence decide where to look.
Keep one shared findings sheet from the start. Give each issue an ID, a screenshot, the page or step, the evidence source and a short description. Figma comments, a spreadsheet or a Notion table all work. What matters is that every line can be traced back to something observed.
Navigation is where visitors decide whether your site is worth their time. Check whether people can tell where they are, where they can go and how to reach what they came for. Washington State University's website usability checklist opens on the same idea: visitors should understand who you are and what you offer, and always know where they are on the site.
Ask five people who do not know your business to find a specific product, price or contact page. Note where they hesitate and what they click first. Even this small sample shows repeated patterns, and a pattern repeated by three of five people is rarely an accident.
Then compare their paths with your analytics. If heatmaps show heavy clicks on a non-clickable image, or internal search is full of terms already in your menu, your labels are not working. Rename first. Restructure only if renaming does not help.
Responsive does not mean usable. A page can resize correctly and still be painful on a phone because of tiny buttons, long forms or heavy images. Test on real devices, not just a narrowed desktop browser, and include at least one older, smaller phone.
If you run a mobile app and not just a site, add checks for permission requests, offline behaviour and back-button behaviour. Our own apps are built with React Native, so one codebase serves both iOS and Android, which means a single fix usually corrects both platforms. Our UAT checklist before launch covers the device and browser coverage you should test before shipping changes.
The checkout is where UX problems cost money directly. Buy something yourself on a phone, using a real card and a real delivery address, and write down every moment of doubt. Then watch recordings of customers who stopped halfway.
If you sell to EU and UK customers, check the compliance side of the experience too. Strong customer authentication under PSD2 can add a bank approval step, so test that the handover to the bank app does not drop the customer back at an empty basket. Make sure consent prompts and privacy notices meet GDPR expectations without blocking the pay button.
Also test failed payments on purpose. A declined card is one of the most common real-life events, and many checkouts handle it badly. Our guide to the discovery phase in app development explains how these edge cases get mapped before any rebuild starts.
Accessibility problems are usability problems for everyone: low contrast hurts on a sunny street, and missing labels hurt anyone in a hurry. The North Carolina Department of Information Technology notes in its accessibility checklist for designers that decisions made during design can prevent defects later in development.
Selling digital services in the EU also brings legal expectations. The European Accessibility Act became enforceable by member states from 28 June 2025, and it covers many consumer-facing digital services, including e-commerce. Treat this audit pass as a first screen, not a compliance certificate. Automated scanners catch only some issues, so a manual keyboard test is essential.
Look for the same thing done in different ways: three button styles, two date formats, inconsistent form errors. These cost users learning time and cost developers maintenance time. A short list of repeated components is often the strongest argument for building a design system during the redesign.
An audit that ends in a 90-item list changes nothing. Give each finding a severity score, then weigh it against the effort to fix it. A simple 0 to 4 scale works: 0 is not a problem, 1 cosmetic, 2 minor, 3 major, and 4 a blocker that stops users completing the task.
| Severity | Meaning | Example | Action |
|---|---|---|---|
| 4 Blocker | Users cannot finish the task | Pay button hidden by a chat widget on mobile | Fix this week |
| 3 Major | Many users struggle or abandon | Forced account creation before payment | Fix in the next sprint |
| 2 Minor | Slows users down but they recover | Unclear label on a filter | Batch with related work |
| 1 Cosmetic | Noticed, rarely matters | Slightly misaligned icon | Fix when passing by |
Suppose a small online shop has 5,000 mobile basket visits a month and 40% of them complete an order, so 2,000 orders. The audit finds that account creation is forced before payment, a severity 3 issue. The team adds guest checkout, and completion rises to 44%.
That is 200 extra orders a month. At an average order of €60, it is about €12,000 in monthly revenue from one change. These numbers are illustrative, not a promise, but the method is real: baseline, finding, severity, expected effect, then measure after release.
Plot each remaining issue on two axes. High impact and low effort goes first. High impact and high effort becomes a planned project. Low impact and high effort usually gets dropped. This one step is where audits earn their fee, because it replaces a long wish list with a sequence.
Turn the scored findings into a short brief: your goal, baseline metrics, the top ten issues with evidence, and the journeys in scope. Fix the cheap items yourself. Take the structural ones to an agency, so quotes cover real problems and are easy to compare.
A good brief also protects your budget. When agencies see evidence, they scope the work against known problems, not assumptions. If you need cost context, our UI/UX design cost breakdown for web apps shows what typically drives the price.
At Axire Infotech, projects follow four steps: Discovery and Planning, Design and Prototyping, Development and Testing, then Launch and Support. A scored audit slots into the first step. It lets the team confirm your goals, test your findings and agree the scope without starting from zero.
Ask any agency how it would validate your audit before designing, how it will measure the result after launch, and who owns each metric. If the answers stay vague, read our other articles on choosing a partner before you sign anything.
One question stays genuinely open: how many test users are enough? Small tests of around five people reliably surface repeated problems, but they cannot size how often a problem occurs. For that you need analytics, and for rare edge cases you may need a larger test. Pick the method by what you need to know.
A focused audit of three to five journeys typically takes a few working days for one reviewer, once analytics and recordings are available. Larger products with many user types take longer. The time is mostly spent gathering evidence and scoring findings, not looking at screens.
Run a full audit before any redesign and then every year or two. Between those, run a light version each quarter, or whenever conversion drops, a major feature ships or you add a new market. Small regular checks catch drift before it becomes expensive.
Yes. A business owner can run most of this checklist with analytics, a phone and five test users. Bring in a specialist when findings point to structural change, when you cannot be objective about your own design, or when accessibility and compliance risk is high.
You need analytics, a session recording or heatmap tool, a spreadsheet or Notion board for findings, and Figma or screenshots for annotations. Add a browser accessibility scanner and a few real phones. None of these needs a large budget.
You do not need to wait for a perfect report before talking to a design team. If your audit points to navigation, mobile or checkout problems, our UI/UX design service can turn the findings into prototypes, and our web development team builds with React, Next.js, Node.js and Supabase. For mobile products, see our cross-platform app development.
Axire Infotech is a development studio based in India with over 3 years of experience, trusted by 10+ companies and working remotely with startups and businesses across Europe. You can browse our projects, including Media Mates, MustQR, Sign Out, Solace Sofas, Nextn and Pehchaan, or see all of our services.
Ready to turn your UX audit checklist into a scoped plan? Contact us with your findings or email hello@axire.in. We respond within 2 hours during business hours.
Let's discuss your project and create something amazing together.
