RoamSEO

Site audits

Crawl your site for SEO issues, read the results, and fix blocked or partial crawls.

A site audit crawls your website and checks every page it reads for SEO issues, such as broken links, missing titles, redirect chains, and pages Google can't index. The crawl itself uses no credits.

Start an audit

  1. Open Site Audit in the project sidebar, under My Site.
  2. Enter the URL to start from. A bare domain like example.com works.
  3. Set Max pages. The default is 50.
  4. Turn on any options you want:
    • Include Lighthouse measures speed and other Lighthouse scores. It tests up to 10 pages, one for each page template it finds, on mobile and desktop.
    • Render JavaScript loads each page in a browser before checking it. Turn it on when your site builds its content with JavaScript and the crawler sees empty pages. It's slower.
  5. To leave part of the site out, open Advanced and list the paths under Exclude paths. See Exclude paths.
  6. Click Start Audit.

The Scan my website button on the project dashboard starts a 50-page audit of your project's site with both options on.

How the crawl works

The crawler starts at your URL, reads robots.txt and your sitemaps, and follows links from page to page. It stays on your site's host: www.example.com and example.com count as the same site, but other subdomains are skipped. It skips pages that robots.txt disallows.

If your site has more pages than Max pages, the audit keeps room for every page in your sitemap. Pages it only finds through links, such as tag archives, fill what's left.

The crawler identifies itself as RoamSEO-Audit. If a firewall or bot protection blocks it, see RoamSEO-Audit crawler for how to allow it.

While an audit runs, the page shows its progress and the pages crawled so far. There's no cancel button; deleting a running audit stops it.

Exclude paths

Use Exclude paths, under Advanced on the audit form, to keep the crawler out of sections you don't want audited, such as a blog archive, internal search results, or cart pages. Enter one path per line, up to 20.

Each line is matched against the start of a URL's path and query string, the same way a Disallow rule in robots.txt is:

You enterWhat it skips
/blog/Every page under /blog/. /blog itself is still crawled.
/blogEvery path that starts with /blog, including /blog-news.
/*/print* matches any characters, so this skips /docs/guide/print and more.
/*?sort=Any URL whose query string starts with sort=.
/*.pdf$$ marks the end of the URL, so this skips only URLs ending in .pdf.

Paths are case-sensitive. The URL the audit starts from can't be excluded.

Excluded pages aren't fetched and don't count toward Max pages, so they don't appear under Pages or Issues. Links that point to them aren't checked, so they're never reported as broken. The orphan page check doesn't run on an audit with excluded paths, because the audit can't see links from the pages it skipped.

The results page lists the paths an audit skipped.

Read the results

  • Issues groups problems by severity: Critical, Warning, and Info. Open an issue to see what it means, how to fix it, and up to 100 affected URLs.
  • Pages lists every page with its status code, title, H1, word count, images, and speed. Filter by Crawl result to find pages that failed, were blocked, or hit a rate limit.
  • Performance appears when Lighthouse ran. Click View issues on a result to see what Lighthouse found on that page.

Use Export to send results to Google Sheets, download a CSV, or download JSON.

Past audits are listed under Previous Audits. Click View to open one. To check your fixes, start a new audit; there's no re-run button.

When an audit looks wrong

  • "No issues found." Before you trust it, open Pages and set Crawl result to Failed or Blocked. If many pages show up, the crawler couldn't read them, so there was nothing to check.
  • "This site's bot protection blocked our crawler." Your firewall or a service like Cloudflare stopped the crawl. Allow the crawler using the steps in RoamSEO-Audit crawler. For a Shopify store, create a signature in Shopify admin under Online Store > Preferences > Crawler access. Then open Project settings, choose the Integrations tab, click Connect on the Shopify card, and paste it.
  • "The crawl stopped early because of the site's rate limit." Your server kept answering with HTTP 429 (Too Many Requests). Allow the crawler, or run a smaller audit.
  • "This audit stopped early after N pages." The results cover the pages it read before stopping.
  • "Crawl stopped early: page limit." This warning under Issues means the audit reached Max pages before it visited every URL it found, so the results cover part of the site. Open it to see how many URLs were left. Start a new audit with a higher limit, or exclude the paths you don't need audited.

Limits

PlanPages per auditAudits running at once
Free505
Paid10,000No limit

Hosted accounts also have a cap on the total size of the audits they keep. If you see "You've reached audit capacity for your account", delete old audits from Previous Audits to make room. Owners and admins can delete audits.

Credits

  • Crawling and checking pages: no credits.
  • Lighthouse: up to about $0.14 per audit.
  • Render JavaScript: the form shows "Costs up to" a dollar amount before you start, about $1.50 per 1,000 pages. You need that much in credits to start. If the cost runs higher than expected, the audit stops.

The dashboard's Scan my website turns both on, so it uses a few cents of credits.

Use your assistant

Ask your agent to run an audit with RoamSEO MCP. It can start one with run_site_audit, check progress, and read the issues, pages and Lighthouse results. From an agent, Lighthouse is off unless you ask for it, and a request over your plan's page limit runs at the limit. It can also pass excludePaths to leave sections out.

The SEO Audit skill runs the audit, reads the results with you, and writes a report with the one to three fixes that matter most. For a checklist to work through by hand, see the technical SEO audit checklist.