Back to blog
What Is a UX Audit, and When Is One Worth Commissioning?
Papi Petrou, SEO & Content Writer Specialist
A UX audit is a structured review of where your website loses people between arriving and completing whatever you wanted them to complete. It is a diagnostic, not a redesign, and the distinction matters because the two are priced and used very differently.
What is a UX audit?
An expert review of the actual journeys on your site, measured against how people behave rather than against a checklist of best practice. The output is a prioritised list of specific problems with specific fixes, ordered by how much each is costing you. What it is not is a redesign proposal. A good audit frequently concludes that the design is fine and the problem is a form field, a missing reassurance, or a page that takes four seconds to become usable on a phone.What does a UX audit include?
A heuristic review of the key journeys, done by someone who has watched a lot of people fail to complete them. Analytics analysis to find where users drop out, which turns opinion into a ranked list. Form and checkout analysis, because that is where abandonment concentrates. Then mobile behaviour reviewed on real devices rather than a resized browser window, accessibility checks covering contrast, keyboard navigation and labelling, and a review of how the site performs when it is slow rather than when it is warm and cached.When is a UX audit worth commissioning?
When traffic is healthy and conversion is not. That combination is the clearest signal the problem is on the site rather than in acquisition, and it is the case where an audit pays for itself fastest. Also before a redesign, which is the moment people least often do it. Redesigning without knowing what currently fails means rebuilding the same problems in a new visual style, and losing the things that quietly worked. It is worth less if traffic is very low, because there is not enough behavioural data to analyse and you are better served by getting visitors first.What kind of problems does an audit usually find?
Forms asking for information the business does not need at that stage. Every extra field costs completions, and phone number is the most expensive one to ask for early. Calls to action that compete. Three equally weighted buttons is the same as none, because the page has not told anyone what to do next. Missing reassurance at the exact moment of doubt: no delivery cost until checkout, no indication of what happens after submitting, no evidence that anyone else has bought. And speed, consistently. Not the marketing homepage, which everyone tests, but the third step of a checkout on a mid range phone on mobile data.How long does it take and what does it cost?
A focused audit of one or two key journeys is typically one to two weeks. A full site review takes longer, and the variable is usually how much analytics work is required rather than how many pages exist. Pricing tracks scope in the same way. What matters more than the number is whether implementation is included or handed back to you as a document, because those are different purchases and they are often quoted as though they were the same one.Should you fix everything the audit finds?
No, and an audit that implies you should is padded. Findings should be ranked by expected impact against effort, and the honest ones say plainly which items are technically true and commercially irrelevant. A well ordered list means the first week of work is the highest value week. That ordering is most of what you are paying for.What happens after the audit?
Implementation, either by your team or by ours, then measurement. The measurement part gets skipped constantly, which is how organisations accumulate changes nobody can prove helped. Where traffic supports it, test the significant changes rather than assuming. Where it does not, implement and monitor the conversion rate over a long enough window to see past the noise. Our UI and UX design team prices implementation separately from the audit so you are not paying a retainer for work that has a natural end point.Doing a first pass yourself
Where audits find the most recoverable value
Checkout and enquiry forms, consistently. They are the narrowest point in the funnel, so a small percentage improvement there is worth more than a large one on a landing page. Navigation on mobile is second. Menus that work on a desktop mouse frequently fail on a thumb, and the failure is silent because the analytics show a visit rather than a frustration. Third is the gap between what an advert promised and what the landing page says. That is usually a coordination problem between two teams rather than a design problem, and it is cheap to fix once someone has noticed it.What a good audit deliverable looks like
Each finding should carry the evidence, the affected pages, the recommended fix, and an honest estimate of expected impact. Screenshots of the actual problem, not stock illustrations of the concept. It should be readable by whoever will implement it. An audit written to impress rather than to be actioned is a document that sits in a folder, and plenty of them do. It should also say what is fine. A list containing only problems gives no sense of proportion, and it tends to trigger changes to things that were working.Running the fixes without breaking what works
Change one significant thing at a time where traffic allows measurement. Bundling six changes into one release means a conversion improvement teaches you nothing about which change caused it, and a decline teaches you even less. Keep a record of what changed and when, so a shift six weeks later can be traced. Most teams cannot answer what changed in the month a metric moved, which turns every analysis into guesswork. Open your own checkout on your own phone, on mobile data, and complete a purchase or an enquiry without using anything you know as the owner. Most people find two or three genuine problems in ten minutes. Then watch one other person do it without helping them. That single exercise consistently produces more useful findings than any amount of reading about web design best practice. It is worth separating an audit from an opinion. Anyone can look at a website and list things they would do differently, and most of that list will be preference. An audit is distinguished by evidence: where people actually leave, what they actually do on a phone, and what the numbers say about the steps between arriving and converting. That distinction also tells you whether you are getting value. If the findings could have been written without opening your analytics, you have bought a design review rather than an audit. Both have their place, and they should not cost the same. The last thing worth saying is that audits age. A site that was reviewed two years ago, has since had a redesign, three campaigns and a checkout change, is a different site. The findings that mattered then have mostly been overtaken, and the ones nobody implemented are usually still sitting there waiting. If you take one thing from this, make it the phone test. Open your own site on your own phone, on mobile data rather than office wifi, and try to complete the thing you want customers to complete. Most owners have never done this on the current version of their own site, and it is the closest thing to a free audit that exists.FAQ
Frequently asked questions
- They overlap heavily. A UX audit looks at the whole experience including usability and accessibility. A CRO audit focuses specifically on the steps that lead to a conversion. In practice most useful audits do both, and the label matters less than the scope.
