Behaviour, Not Taste
Where people stop, on which template, on which device.
Website Design Audit
Website Design Audit That Measures Instead of Offering Opinions
A design audit that lists what the reviewer would have done differently is an opinion with a report cover. A website design audit should measure: where visitors actually stop, what is slow and by how much, what an unbriefed person understands from the page, and which of those findings is worth acting on first.
Every Finding Has Evidence Behind It
A finding without a measurement is a preference, and preferences do not justify a budget.
Performance Measured
Field data from real visitors, not only a lab score.
Comprehension Tested
What a first-time visitor takes from the page.
Ranked by Impact
Thirty findings of equal weight is not a report, it is a list.
Actionable
What to change, where, and what to expect from it.
What a website design audit inspects
An audit is only worth commissioning if it produces findings somebody can act on. Impressions do not qualify. Each item below produces a specific result — a location, a measurement, or a documented failure that can be verified by repeating the check.
Primary action clarityWhether the intended next step is obvious on each key page
Message hierarchyWhether the most important thing is the most prominent thing
Above-fold comprehensionWhether a stranger can tell what the business does without scrolling
Navigation labellingWhether menu labels match what visitors expect behind them
Mobile parityWhether every task completable on desktop is completable on a phone
Form frictionField count, validation behaviour and mobile input types
Trust signal placementWhether evidence appears where doubt occurs, not only at the top
Accessibility conformanceContrast, focus order, keyboard operation, heading structure
Layout stabilityElements shifting after load, and what causes it
Load behaviourWhat renders first and how long the page appears empty
Content legibilityLine length, type size and spacing at real device widths
ConsistencyWhether the same component behaves the same way across the site
Above-fold comprehension is the check that most often surprises the people who commissioned the site. Teams know what their business does and cannot un-know it, so the first screen reads as obvious to them and as ambiguous to everyone else.
Audit, Explained
What Gets Measured?
Four things, and each produces evidence rather than an opinion.
Request an Audit →- 1
Behaviour
Drop-off by step, template and device, from your own analytics.
- 2
Performance
Load and interaction, in the field and in the lab.
- 3
Comprehension
What people understand, tested rather than assumed.
- 4
Accessibility
Keyboard, contrast and structure, checked manually as well as automatically.
How We Audit
Measure, test with people, rank, report.
Gather Data
Analytics, field performance and search position.
Test With People
Real tasks, watched, on real devices.
Check Access
Keyboard and structure, manually.
Rank Findings
By expected impact, not by how easy they are to write up.
Report
What to change, where, and why it is ranked there.
When an audit is the right first step
An audit is diagnostic. It earns its cost when the problem is not yet defined, and it delays the work when the problem is already known.
Sites that get traffic and few enquiries
The clearest case. Visitors are arriving and leaving, and nobody can say at which point. The audit converts a general concern into a list of specific, addressable failures.
Businesses considering a rebuild
The most valuable timing and the most often skipped. A rebuild without a diagnosis carries every existing problem into a new visual layer and risks discarding whatever was working.
Sites built some years ago
Where the site predates the current offer, the current audience, or the current share of mobile traffic. Frequently the structure is sound and specific pages have drifted out of date.
Businesses inheriting a site
New marketing leadership starts with the previous team's account of things. An audit substitutes an evidenced baseline, which is faster and less political than reconstructing the history.
Sites facing an accessibility requirement
Where a complaint, a procurement request or an obligation has made conformance urgent. This needs a technical review alongside the usability findings, and the applicable standard should be confirmed rather than assumed — VALIDATION REQUIRED.
Method
Why Unranked Findings Are Not Useful
A list of thirty problems presented as equals leaves the reader to guess where to start.
What is wrong with a long list?
Audits are frequently judged on length, so they include everything findable — which mixes a broken mobile checkout with an inconsistent heading margin.
Presented at the same weight, the reader works from the top, and the top is usually ordered by section rather than by consequence.
Ranking forces the harder judgement: which two or three findings, if fixed, would change the outcome. That is a shorter and considerably more useful report.
Why test with people rather than review?
Because an experienced reviewer knows what the page means and cannot un-know it. They can identify mechanical problems and cannot judge comprehension.
Watching five people attempt a real task on the site surfaces things no review does — the label read as something else, the step nobody notices, the assumption the page makes about prior knowledge.
It is not expensive and it converts opinion into observation, which is the difference between a report that gets argued about and one that gets acted on.
Findings without priority are not a deliverable
An audit returning forty observations has moved the work rather than done it. The client now has to decide what matters, which is the judgement they were paying for. This is why so many commissioned audits are read once and filed.
Useful ranking combines two independent things: how much the problem costs, and how much fixing it costs. A primary call to action that is invisible on mobile is severe and cheap to fix, so it goes first. A recommendation to restructure the entire navigation may be correct and belongs in a planned project rather than a quick-wins list.
The category needing most discipline is cheap and inconsequential — a slightly inconsistent margin, a heading one size too small. These are satisfying to fix and change nothing, and a team working from an unsorted list will spend its energy there because the items are easy.
What an audit can establish, and what it cannot
An audit can establish what the site does: what renders, what is reachable, where forms fail, which contrast pairings fall short, what shifts after load. All of it is observable and verifiable by anyone repeating the check.
It cannot establish why a specific visitor did not convert. That requires either analytics showing where people stop, or watching real users attempt the task. An audit that asserts a motive for visitor behaviour is offering a hypothesis in the language of a finding.
It also cannot tell you whether the offer is right, the pricing is competitive, or the traffic is well-targeted. A site can be well designed and underperform because the visitors arriving were never going to buy — which is a different problem, and worth ruling in or out before redesigning anything.
Where an audit sits relative to a rebuild
The common sequence is to decide on a rebuild and then design it. The more economical sequence is to audit first, because the findings become the brief. A rebuild informed by specific, evidenced failures produces a better result than one driven by a general sense that the site feels dated.
It also frequently reduces the scope. Many sites that feel like they need replacing have a sound structure and three or four pages doing real damage. Establishing that saves the cost of a rebuild and delivers most of the benefit, which is a finding worth having before the budget is committed.
Where a rebuild genuinely is warranted, the audit still pays for itself by defining what must be preserved. That matters for search visibility as much as for design, which is why an audit of this kind runs alongside the web design planning rather than after it, and why the URL and content inventory belongs in the same conversation.
Questions, answered.
A website design audit measures how a site performs — behaviour, speed, comprehension and accessibility — and ranks the findings by expected impact rather than listing them equally.
Every finding has a measurement behind it. A reviewer's preferences are not evidence, and they do not justify a budget or survive an internal argument.
Where the scope allows, yes. An experienced reviewer knows what the page means and cannot judge comprehension; watching five people attempt a task surfaces what review cannot.
Fewer than most audits produce, and ranked. The useful output is which two or three changes would alter the outcome, not a list of everything findable.
Partly — structure and performance overlap. For a full examination of search visibility see SEO audit.
A design audit examines whether visitors can understand the site and complete what they came to do. An SEO audit examines whether search engines can crawl, index and interpret it. They overlap at performance, mobile and structure, and they answer different questions — a site can pass one comfortably and fail the other.
Only if the evidence supports it, and often it does not. Many sites have a sound structure and a small number of pages causing most of the loss. An audit concluding that three pages need work rather than the whole site is a legitimate finding, and it saves considerably more than it costs.
Findings with locations attached so each can be verified independently, evidence for each — a screenshot, a measurement, an analytics view — and a prioritisation separating what is urgent and cheap from what is structural and expensive. A finding that says the navigation is confusing is not actionable; one naming the labels and the paths visitors took instead is.
It is not required and it substantially improves the result. Analytics shows where visitors actually stop, which turns a list of possible problems into a ranked list of real ones. Without it, findings are prioritised on severity and reach rather than on observed behaviour.
Parts of it. Checking contrast, testing forms on a real phone and attempting your own primary task on a mobile connection all find genuine problems and cost nothing. What is harder internally is judging comprehension, because everyone on the team already knows what the business does and cannot read the first screen as a stranger would.
Still deciding if website design audit is right for you?
Talk to UsThirty Findings, All the Same Size
Audit reports are usually judged by weight. A long document feels thorough, and a short one feels like it missed something, so audits include everything that can be found.
The result puts a broken mobile checkout on the same footing as an inconsistent margin. Both are findings, both are described in a paragraph, and the reader has no way to tell which one is costing money.
Deciding which three findings actually matter is harder than listing thirty, and it is the only part of the exercise the client can act on.
Free Initial Review
Find Out What Your Site Is Actually Doing
Send us your site and access to your analytics. We will measure where visitors stop and rank what is worth changing.
Claim Your Free Marketing Audit
Enhance Your Brand Potential At No Cost!
- Expect a response within 24 hours
- NDA available upon request
- Dedicated product specialists