The Redirects tab lists the pages of your site that answer 404 or 410, most linked first, each with a suggested new address. Check the suggestions, change the ones you want, and download the redirects as a file for your server or platform. Nothing changes on your site until you publish that file.
Open Site Crawler, then Redirects, and pick a site from My Sites. The list comes from the site's Site Pages, so it is up to date after every scan. Viewing it is on every plan and costs no credits.
How a new address is suggested
For each old URL, PerfBee looks at the pages of the same site that work, are found by the latest scan and are not noindex or canonicalised to another page, in this order:
- Same page name: a live page with the same last part of the URL, for example after the page moved to another folder, or gained or lost a trailing slash or
.html. - Similar name: in the same first folder, the live page whose name shares the most words, as long as it keeps every model code of the old one (a word with a digit, such as
9462). Another model is never suggested. - Parent folder: the nearest live folder above the old URL, such as
/dishwashers/.
The homepage is never suggested: Google treats a redirect to the homepage as a soft 404, so the old page's value is lost anyway. A URL in a language folder (such as /uk-en/ or /de/) is only matched inside that folder. When nothing fits, the row waits for a target of your own.
Broken links that are not moved pages
Some broken URLs come from a mistake in the link itself, such as /tv/void(0), a template placeholder like {{url}}, or a link written without https:// so its domain became a folder. A redirect would hide the broken link, so these are listed above the table instead, grouped by the part that gives them away. Fix the link where it sits: the Fixes tab shows the element.
Change, leave out, reset
- Type a path (
/new-page) or a full URL in the New address box. It is saved when you leave the box. - Untick Include to leave a URL out of the download, for example a page you removed on purpose.
- Use suggestion or Reset forgets your choice and brings the suggestion back.
Your choices are kept for the old URL, so they survive the next scan. Team members can change them; viewers can only look.
Download
The Export menu downloads the rows that are ready (a new address, included), on Starter and up:
- Apache (.htaccess):
RewriteRulelines. Put them at the top of.htaccess, above any CMS block such as WordPress's, or in the site's<VirtualHost>. - Nginx: one
location =line per redirect, for the site'sserver { }block. Reload Nginx afterwards. - Cloudflare Bulk Redirects: a CSV to upload to a Bulk Redirect List.
- Shopify (CSV): the file Shopify's URL redirects import expects (Redirect from, Redirect to).
- CSV and Excel: old URL, new URL, how it was matched, status code and links to it.
Every redirect is a 301. An old URL with a query string, or with characters a server rule cannot hold safely, is left out of the server files and listed at the end of the Apache and Nginx files.
Check published redirects
Once the redirects are live, click Check published redirects. A fast scan reads each ready old URL once, up to 2,000, and records where it lands. It is billed like a fast scan, for the URLs it reads, and the confirmation shows the credits first. The result appears when the scan has ended and Site Pages has taken it in:
- Works: redirects and lands on the address you set, and that page answers.
- Goes elsewhere: redirects to another page, or to one that fails.
- Still 404 / 410: the redirect is not live yet.
- Answers by itself: the old URL works again without a redirect.
- Not read: the scan could not reach it, for example the site refused PerfBeeBot.
While it runs, the check is in the Scans list, where you can follow or cancel it; when it ends, its result is on the Redirects tab. It does not count as the site's latest scan, and another scan of the site can start once it has finished.