# Broken links and redirect checks

Scheduled crawls, issue types, fixed issues, alerts and limits.

Broken links works on the website picked in the header switcher. Open Dashboard → Broken links to see the latest crawl, the open issues and the crawl history.

## What's checked

- Pages on your site that answer 4xx or 5xx, or don't answer, listed with each page that links to them.
- Links to other sites that are gone. Answers that mean a site refused an automated request or asked it to slow down (401, 403, 429) aren't treated as broken.
- Images that fail to load, on your site or elsewhere.
- Redirect chains (a link that takes more than one redirect to land) and redirect loops.

## How the crawl works

We start at your home page and follow your own links breadth-first with the SerpfullAuditor user agent, respecting robots.txt. Links to other sites and images get a light HEAD request, falling back to GET when a server doesn't support HEAD. Forms are never submitted and nothing is clicked.

## Schedule and manual crawls

Each website is crawled weekly by default. Choose daily (from Growth), weekly or off on the Broken links page, and press Run crawl now after a release. Manual crawls have a daily allowance and a 15-minute pause between them.

## Fixed issues

An issue closes when the page it was found on is crawled again and the problem has gone. Issues on pages a crawl didn't reach stay open until a crawl reaches them.

## Limits

- Pages crawled per crawl: Starter 200, Growth 500, Scale 1,500, Pro 5,000, Enterprise 15,000.
- Links and images checked per crawl: the same numbers.
- Manual crawls per day: 1, 2, 4, 10 and 20.
- Crawl history kept: 30, 90, 180, 365 and 730 days.

## Alerts and webhooks

broken_links.crawl.completed (every finished crawl, with counts) and broken_links.issue.new (new issues found, with up to 50 of them). Use them as Workflow triggers or add them in Integrations → Webhooks. New issues also reach the notification bell unless you switch alerts off for the site.