Ask SEOs how long to keep redirects, and many will say “forever”. I’m in that same camp. Most sites have more pressing issues to focus on, and you don’t want to throw away any redirects that still have value.
But sometimes “forever” isn’t always for the best. A growing list of redirects can cause problems ranging from performance issues to long redirect chains that land a user on a page that bears little resemblance to the original page they were looking for.
The web is always evolving and changing. Page changes and deletions happen more than you think. Websites go through redesigns, rebranding, and technology changes which all can contribute to changing and discarded URLs.
A site migration, especially if you have to port over the old redirects, is a good time to evaluate whether old redirects are worth keeping.
Sometimes you’ve just got to wade in and clean up.
How long should you keep old redirects?
Most of you will know what a redirect is, but as a quick definition, a redirect will send a user (and search engine bots) to a different URL than the one they requested. There are several different types of redirects, but in this article, we will be primarily discussing the HTTP redirects, namely the 301 (permanent) and 302 (temporary) redirects that exist in a list somewhere in your system.
If you have decided to cull your list of redirects, you might be wondering how long Google recommends keeping redirects. Surprisingly the answer is just 1 year, which doesn’t seem long enough.
Remember though, Organic Search is not the only way that people find your site, people do bookmark URLs and you don’t want to throw away referral traffic if you can help it. Despite the 1 year guideline from Google, I’m still reluctant to delete a redirect that has a decent backlink associated with it and is still sending traffic to the site.
As a SEO Consultant, I’ve often looked at redirect lists that have no documentation and were implemented by staff that are long gone. So when I curate old redirects, I use a data based approach rather than an arbitrary timeline to decide what redirect stays and what gets deleted.
Specifically I look for redirects where the source URL doesn’t have traffic, nor a good backlink and is not known/crawled by Google.
How to evaluate an old redirect list
Here’s how I go about evaluating an old redirect list to see if I can kill any redirects that are not needed anymore.
Step 1: Build your old redirect evaluation list
A redirect has a source URL and a target URL. The source URL is the URL that needs to be redirected for whatever reason, the target URL is the destination page where we would like the user to end up.
In this exercise we are only going to look at the source URLs to determine whether it is still necessary to have a redirect in place.
To get your list, you’ll need to know how redirects are implemented on your site.
- Redirects could be handled by your web server; in which case you’ll need to find the configuration file that contains the redirect instructions.
- Or you may find that your redirects are handled at the application level, such as by a WordPress plugin.
Once you find the redirects, either export or copy the redirect list to a spreadsheet. The source URL is usually the first URL listed per redirect.
In the steps below we will be cross referencing them against the lists you obtain in the steps below. If you want to take this evaluation to the next level, you can also check the indexation status of the source URL (explained in Step 4 below).
If the URL is unknown to Google, doesn’t have traffic or backlinks, then it is completely safe to delete the redirect associated with it.
A note about wildcards
One complicating factor is if you have wildcard redirects. A wildcard redirect is a type of redirect that uses a wildcard character (usually an asterisk *) to capture and redirect multiple source URLs that fit a specific pattern to a target URL. If there aren’t too many, you can search for the pattern in your traffic, backlink and indexation lists to see if you find any URL that would be in scope.
Do you need your redirect for an internal link?
Before we get into the steps, I should mention that I’m assuming you don’t need any of the redirects you are evaluating for the internal links on your site.
If you are not sure, run a crawl for your site. Get a list of the redirects the crawler found and compare this to your evaluation list. If you find a match, you have two choices:
- Fix the internal link to point directly to the target URL.
- Remove the redirect from your evaluation list, we don’t want to introduce new 404s into your site!
Now that we have a list, let’s move forward.
Step 2: Check your traffic
If any of your source URLs are getting significant traffic, then you should keep the redirect. I like to cross reference my list against 1 year of GA4 landing page data, but you could also look at Google Search Console data. Depending on the GA4 setup, the data can be messy due to URLs with tracking parameters, so you may want to use the GSC Performance Report for cleaner data. I prefer GA4 if possible because it includes traffic data from all channels not just organic.
Both GSC and GA4 have row limits that impact exports which is annoying if you have a larger site. I have an article that shows you several ways to get around the GSC 1,000 row limit. For GA4, check the standard Landing Page report to see if you have more than 250 rows for the 1 year timeframe. Unfortunately, you will only be able to export 250 rows from this report. If that is a problem for you, you can create a GA4 exploration using landing pages as the metric and export from there.
Once you have your export, compare the exported URLs against your source URL list using an Excel tool like vlookup().
If the export is from GA4, you may need to add the protocol and the domain to your exported URL, to match up with your source URLs list. You also may want to decide on how many Clicks or Sessions are actually “significant” traffic.
URLs from your evaluation list that show up in the traffic report are “keep” candidates.
Step 3: Check your backlinks
Backlinks are hard to get, why would you throw any away? If a backlink results in a 404 (Not Found error) the SEO authority from that site will not benefit yours. I alway recommend fixing broken backlinks unless the linking site is obvious spam or otherwise low quality.
So let’s figure out which of the URLs on your evaluation list have a backlink.
If you have access to tools like Majestic, aHrefs or SEMrush, you can export a list of backlinks for your site. You may want to take the extra step of filtering by domain authority if there are a lot of low quality sites in your backlinks. You’ll want the target URLs of the backlinks so copy those from the export and then sort and deduplicate before cross referencing them with your evaluation list.
Alternatively, you can get a list of URLs from your site that have backlinks from Google Search Console. Find the links menu item and then click on the “More” link for the Top linked pages in the External Links box.
Step 4: Does Google still crawl the URL?
This next step is optional, but I like to be conservative and do it. I think it’s worth checking whether Google still crawls the source URL of your redirect as a final check in determining whether to keep it.
The best way to check for this is to get a copy of your server logs (you want the access logs) and filter this for Googlebot accesses. If your source URL doesn’t show up in the server logs that’s a good sign that Google doesn’t care about the URL and the redirect is not needed.
Since, many SEOs have trouble getting access to the server logs, as an alternative you can check indexation data using the GSC API. Screaming Frog makes this easy (you’ll need the paid version) via their API integration.
- Configure the GSC API integration in Screaming Frog, making sure to check the “Enable URL Inspection” checkbox under the URL Inspection tab.
- You’ll want a browser handy that is logged into the same Google account that has access to the GSC for your site so that the authentication step works as you configure the API integration.
- Upload the list of source URLs on your evaluation list using the List Mode setting of Screaming Frog and crawl them.
- Navigate to the “Search Console” tab and use the Export button to export your data.
- In the spreadsheet, look for the “Summary” and “Coverage” columns
Note that you are limited to 2,000 API requests per day for GSC.
If your URL is “Unknown to Google, then you don’t need the redirect.
If the URL is “not on Google” due to a redirect, you can delete the redirect if you didn’t find a backlink or significant traffic for the source URL in Step 2 or Step 3.
Step 5: Do the redirects go somewhere useful?
At this point, you should have a list of redirects you want to keep. The question is, are they useful? Your redirects may point to pages that 404 or maybe you are redirecting a specific page to a general category page.
Regarding the latter, I rarely see organic traffic that used to go to a specific page get moved over to a general page. It’s just not a good user experience. As much as possible a redirect should go to a similar page. If you don’t have a similar page to redirect to, either create one or remove the redirect and let it be a 404.
To find 404s or redirect chains, you can upload the list of your target URLs to Screaming Frog when it is in “List” mode. Ideally you want to see all URLs you uploaded return a 200.
With this final step in the analysis done, you can now remove the redirects that are not useful, and don’t satisfy one of the above criteria (have traffic, a good backlink and is known by Google).
Evaluating old redirects may not be on top of your list of technical debt issues you need to fix, but I’ve definitely discovered some problems (redirect chains, redirects to 404s, poor choice of a target URL) when I’ve gone through this exercise.
Accordingly, for a site migration project, I will usually include an old redirect evaluation task on my checklist so I’m not porting over problems or unnecessary redirects.
