A real SEO audit.
Without the price tag.

What a real audit covers, and why issue counts mislead

Most audit reports open with a number: 247 issues found. It is the least useful sentence in the document. Issue counts are dominated by whatever the crawler found most of, which is usually a trivial problem repeated across every page, and they make a site with one catastrophic problem look healthier than a site with two hundred cosmetic ones.

A useful audit separates four layers, because each fails differently and each is fixed by different people.

Crawlability and indexation. Can search engines reach the page, and are you permitting them to keep it? This layer is binary and it outranks everything below it. A page that cannot be crawled does not benefit from any amount of content work, so a single misplaced noindex on a template is worth more attention than every missing alt attribute on the site combined.

Technical health. Status codes, redirect behaviour, canonical consistency, structured data validity, sitemap and robots agreement. These are the problems that quietly cap what the rest of your work can achieve.

On-page. Titles, meta descriptions, heading hierarchy, internal linking, content depth against what the query actually wants. This is where most of the addressable upside sits for most sites.

Experience. Performance, mobile rendering, accessibility, layout stability. Weaker ranking signals than the layers above, but they affect whether the traffic you earn does anything useful once it arrives.

Severity is a judgement, not a measurement

No two tools agree on severity, because severity is contextual. A missing meta description is trivial on a page nobody lands on and material on a page earning thousands of impressions. Duplicate title tags matter enormously across a product catalogue and not at all across three legal pages. Treat any severity label as a starting suggestion to be checked against your own traffic, not a verdict.

The practical test of an audit is whether you can act on it on the day you receive it. If it needs a specialist to interpret before anyone can start, it has moved the problem rather than solved it.

How UtilitySEO handles this

The audit crawls every page on your site, up to 10,000 on Enterprise, renders JavaScript and runs 100+ checks on each page, on mobile and desktop. It then adds what your Search Console, GA4 and PageSpeed data say, so the results come back as one list of fixes in order of importance, each with a plain English guide to fixing it.

Work through that list as a checklist, add fixes to your to-do list, or export the full audit as a CSV for your own spreadsheet. PDF reports are there too, white-label on Enterprise. Re-scan any time to check a fix worked.

It starts at £39/mo on Minimum, with no per-feature add-ons, and we help you connect Search Console and Analytics and set up your first project for free. Not sure you need paid software at all? Read our honest comparison of SEO software and Google's free tools.

The same tool under five different names

People search for this category under names that barely overlap, which is worth saying plainly because it causes real confusion when comparing products.

An SEO analysis app, an SEO measurement tool, an SEO health tool, an SEO diagnostic tool and an SEO verification tool are, in practice, the same purchase: something that crawls your site, checks it against what search engines need, and tells you what is wrong. The names come from which part of the job the vendor decided to lead with.

The differences that actually matter are narrower than the naming suggests:

Does it render JavaScript? A crawler that reads raw HTML on a React site reports a blank page and calls it a content problem.

Does it rank the findings by anything real? Four hundred issues sorted by a severity label somebody chose is a list nobody finishes. Sorted by which pages actually earn impressions, it becomes a morning's work.

Does it check your own data, or a simulation? A verification run against Search Console tells you what Google did. A simulated audit tells you what a crawler thinks Google might do.

"Most effective" and "top 10" lists, briefly

Searching for the most effective SEO software or a top 10 SEO analysis tools list mostly returns affiliate rankings, and the ordering usually reflects commission rather than fit. The more useful filter is the three questions above, plus one more: what happens on the day you find four hundred issues and have two hours. That is the day the tool either helps or becomes a tab you stop opening.

Duplicate content: internal, external, and which one matters

A duplicate content tool is really two checks wearing one name, and they have different consequences.

Internal duplication is your own pages competing with each other: the same guide published twice, a printer-friendly variant, a paginated archive repeating its intro, parameters generating a hundred near-identical URLs. This is the one that actually costs you, because Google picks one version and your signals split across the rest. It is also the one you can fix in an afternoon with a canonical or a redirect.

External duplication is someone else publishing your text. This matters far less than people fear. Google is generally good at identifying the original, and an external duplicate content checker mostly returns scrapers republishing your feed, which need no action at all.

The check worth running regularly is the internal one. Two pages of yours in the same result set is a problem you created and can therefore solve.

What a duplicate content check should actually report

Not a similarity percentage. Two pages can share 80% of their words and be entirely correct, if the shared part is boilerplate. What you want is which of your own URLs Google is treating as alternatives for the same query, because that is the split that costs impressions.