Run a site audit

Site Audit crawls your website the way a search engine does and lists the technical problems that can hold your rankings back — broken links, missing or badly sized titles, thin pages, images without alt text and more. Each problem comes with an explanation and the affected pages, so you know what to fix and where.

Webflow sites: Site Audit isn't available for a site connected through the ProjectHQ App for Webflow. ProjectHQ reads Webflow sites only through Webflow's official API and never crawls them, and links pointing to Webflow sites aren't checked either. See Publish to Webflow.

When the audit runs

ProjectHQ crawls your site automatically twice a month. Your first audit runs as soon as you set up your project; until it has finished, the page says No site audit data available yet.

To crawl again after you've fixed something, click Rescan website in the bar on the right (before the first audit it reads Run site audit) and confirm with Start scan. While it runs, the button shows the progress, for example Crawling pages… 30%. You can start one manual scan every 24 hours; the button tells you when the next one is possible.

Read the results

1. Site Health — a score from 0 to 100 for the whole site, with the date of the last crawl.
2. Issues Distribution — how many issues are Errors (fix these first), Warnings and Notices. Click one to filter the list.
3. URLs — how many pages were crawled and how many of them have errors.
4. All Levels, Errors, Warnings, Notices — filter the issue list.
5. Ignore — the crossed-out eye hides an issue type you don't care about (see below).
6. Rescan website.
7. Settings — ignore rules and URL blacklisting.

Each card shows how many pages have the issue, its category (such as Links, HTML tags or Content quality) and what to do about it.

Fix an issue

Click an issue card to see every affected page:

• The table shows the data that matters for this issue — a title's length, the broken link's target, a page's word count. Use the URL and Title filters to narrow it down, and click a page to open its full details (the same view as in Webpages).
• Copy AI prompt (1) copies a ready-made prompt you can paste into an AI coding assistant. For issues it can fix in your code, the prompt asks it to fix them, with the affected URLs included; for the rest, it asks for the likely cause and options to choose from.

After you've fixed the pages, run a rescan to clear the issue.

Thin content

Thin content isn't a simple word count. ProjectHQ's AI reads each page and judges it against what that kind of page should deliver — a short pricing page is fine, an article that barely covers its topic isn't. The Why It's Thin column (2) explains what's missing on each page, and Page Type shows how the page was classified.

Ignore what doesn't apply

Some issues are deliberate — a login page that's meant to be short, a redirect you want. Ignored issues no longer count towards your results.

• A whole issue type — click the crossed-out eye on its card and confirm with Ignore Issues.
• One page — on the issue's page list, use the two icons at the end of a row: Ignore this issue on this page (the crossed-out eye) hides just this issue for that page, and Ignore all issues on this page (the circle with a line through it) hides every issue on that page. Hover over an icon to see which is which; the confirmation tells you too.

To see or undo what you've ignored, click Settings in the bar on the right. It opens Ignore rules in the project settings, with each rule and a bin to remove it.

Leave parts of your site out

In the same settings, URL blacklisting tells the crawler what to skip. Changes save automatically and apply from the next crawl.

Type a value and press Enter or click +; click × to remove it.

Tip: Problems the audit finds also feed Insights, next to the other improvements worth your time.