How a review runs, start to finish.

The crawler walks the site the way a search engine would, following links until it has the full list of pages a visitor can actually reach.

Most problems do not live on the homepage. A broken booking form, a page served without HTTPS, a template that collapses on mobile: these hide two or three clicks in, on the pages that convert.

Every page runs through the same set of checks: load performance on a throttled connection, TLS and security headers, mixed content, broken links and redirect chains, mobile rendering, form behaviour, metadata and heading structure.

The server itself is checked too: exposed files, stale software, and anything left behind from a past build. Because this is automated, it runs on every page rather than a sample of three.

Raw scanner output is mostly noise. A large share of what any tool reports is a false positive, a duplicate, or something that changes nothing for a visitor.

So every flag is re-tested against the live page and has to produce evidence: the failing request, the response header, the screenshot at the width where the layout breaks. If it cannot be reproduced, it is dropped before anyone sees it. We read the surviving list before the report is sent.

The report is written for the person who owns the business, not for a developer. Each finding names the page it happens on, describes the problem in one or two sentences, and says plainly what it does to a visitor or to search ranking.

Where a problem repeats across the site, we say how many pages carry it and give examples. You get this whether or not you go further with us.

If you want the work done, we quote a fixed price for it after the audit, once we know how much of the site is affected. Small, contained problems are repaired in place. Where the template or the build is the root cause, we rebuild those parts properly.

We do not describe this work as quick or minor. The audit tells us the real scope, and the price reflects it.