Sitewide SEO Audit
Crawl a site and score every page
What Is the Sitewide SEO Audit?
Most SEO problems live between pages, not on them: forty products sharing one title, a template that never got a meta description, a page nothing links to. The Sitewide SEO Audit crawls up to 250 pages of a site, checks each one for 20 common problems, and turns the result into a scored report you can work down, with issues grouped by type and the worst pages first.
Key Features
How the Score Is Worked Out
Each check has a weight based on how much it matters, and costs that weight times the share of crawled pages it affects. A missing title on every page costs the full ten points; on one page in fifty it costs a fifth of a point. So the score moves when you fix something widespread and barely notices a single stray page. The "Where the points go" chart shows exactly where the points went, which makes the first fix obvious.
Fix Templates, Not Pages
Duplicate titles, missing descriptions and thin pages usually come from one template rather than forty separate mistakes. Grouping findings by type makes that visible: one row with forty pages under it is one job for a developer, not forty tickets. Work down the issues from the most expensive, then use the worst-pages table to check the stragglers.
What a Crawl Can and Can't See
The crawler reads pages the way a search engine first does, as plain HTML, and it follows robots.txt. Pages reachable only through JavaScript or forms won't be found. On a site bigger than your page limit, the report describes the pages it read, says so, and gives the site's size from its sitemap. Run 250 pages for the fullest picture.
Common Use Cases
- Auditing a client site before quoting SEO work
- Checking a redesign or migration for errors and lost titles
- Finding the template bugs that affect the most pages
- Seeing whether last month's fixes moved the score
- Handing developers a prioritized list of pages to fix
Frequently Asked Questions
Is the Sitewide SEO Audit free?
Yes. It's free with a ProxyHorizon account, and every account gets 25 runs a day shared across all our data tools. You can read about the tool without signing in, but running it needs an account because each run fetches live data.
What does it check?
Pages returning errors, noindex, canonicals pointing elsewhere, missing and duplicate titles, title length, missing and duplicate meta descriptions, description length, missing or multiple H1s, thin content under 300 words, links to broken pages and to redirects, pages with one internal link or none, pages four or more clicks deep, images without alt text, slow server responses, missing canonical tags and missing language attributes.
How many pages does it crawl?
25, 50, 100 or 250, your choice. It starts at the URL you enter and follows internal links outward, so the pages closest to the start are read first. On a site bigger than your limit, the report says how many pages it read and how many the site has at least, and describes those pages rather than the whole site. A run has about two and a half minutes to crawl; on a slow site it stops where it got to and says so.
Does the crawler follow robots.txt?
Yes. It skips anything robots.txt disallows for all crawlers and slows down when the file sets a crawl delay. It requests pages directly, which is how search engines first see them. If a site blocks plain requests outright, it loads up to 100 pages in a real browser instead and the report says so.
Do you keep the crawl?
For 24 hours. The crawl behind a report is stored for a day so the other sitewide tools can reuse it for the same site and page limit, instantly and without crawling again. It only holds public pages, so anyone checking the same site that day gets the same crawl. Choose "Crawl again now" whenever you want fresh data. We also log which tool you ran and when, to apply the daily limit.
Crawling whole sites for SEO at scale?
Proxies for SEO monitoring