The question of what to do with 404 errors on your website is one of those eternal conversations for those who work in SEO. Clients and bosses can often live in mortal fear of their users seeing a 404 page.
For many, the “solution” is to blanket redirect users from any page returning a 404 status to their homepage.
At first glance it seems like there are compelling reasons for this:
- It maintains link juice passed by backlinks
- Users avoid 404 errors
- It keeps Google from seeing 404 errors
However, even Google itself has come out advising against this practice.
Despite this, 11.4% of redirected backlinks send users to the the homepage.
So what exactly makes this such a bad idea?
They can confuse Google
As we mentioned above, many people like to redirect to the homepage or a category page in an effort to preserve the value passed by links. But if you watch that Google Webmaster Central Hangout with John Mueller, John specifically states that category page would end up being interpreted as a soft 404:
And here’s another Webmaster Central Hangout a few months later where he specifically mentions redirects to homepages and category pages as practices to avoid:
Watch the second video to the 9-minute mark and you’ll get the part specifically about preserving link juice.
“We’ll see that as a soft 404 and essentially treat it as a 404 anyways.”
Even though you’re using 301 redirects — which normally pass full link juice — you’ll lose all of the original page’s ranking signals if you institute blanket redirects to avoid 404s.
Now that’s bad enough for a category page, but redirecting unavailable URLs to your homepage could also look like you’re trying to confuse or cheat Google by sneakily passing ranking power to your homepage.
Glenn Gabe at G-Squared Interactive published a case study in July of 2016 in which he used blanket redirects during a site migration. The result was a rise in soft 404s in Google Search Console as well as a loss of ranking for the redirected pages in SERPs.
All the effort to maintain rankings came to nothing.
And this makes a lot of sense for Google to treat these redirects this way because…
They confuse users
Imagine you’re a user who clicks on a link while reading a web page. You have certain expectations of what information and content the page will contain. And you click on the link with a certain intent: to learn something or to complete a certain action (buy a product, sign up for a webinar, etc.).
But when you click the link, you wind up on that site’s homepage or a category or tag page.
As a user this is pretty frustrating. You thought you were getting one thing, but you wound up with something completely different. And you have no idea why.
Let’s look at a real life example (taken from the WooRank research mentioned in the intro).
Here’s an Ikea product someone pinned to their Pinterest board:
<span style=”text-align: center; display: block”>
<a data-pin-do=”embedPin” href=”https://www.pinterest.com/pin/516647388474841471/”></a>
If I were in the market for a cabinet, my first instinct would be to click through to the Ikea site so I could learn more about the product specs and options — dimensions, colors available, etc. But that product’s not available anymore, so when you click on it, you wind up on Ikea’s homepage.
Now as a user, I don’t know why I’m on Ikea’s homepage. I don’t know the product isn’t available, so I have to go searching through the site in an attempt to find it.
They can confuse webmasters
Finally, redirecting all 404s to your homepage will hide issues and problems from the website owner.
Here are some potential problems you would miss by redirecting all 404s to another valid page:
- Someone added a link to one of your pages, but mistyped the URL. You could be missing out on some great links without even knowing it.
- Someone took down a page, but people are still trying to access it via bookmarks or external links.
- A search engine could consider your homepage as a substitute for the removed page’s content, causing it to potentially rank lower (or not at all) for your main keyword.
All 3 of these issues can cost you visitors, rankings, leads and, ultimately, sales.
It’s also worth noting that, for particularly large websites that can have listings or user-generated content come and go, redirecting all 404s will result in your server processing hundreds (or even thousands) of redirects per day. More than it would otherwise need to, putting an unnecessary strain on your infrastructure and potentially slowing load times for users.
Use redirects intelligently and let the rest 404
The best way to handle 404s is to set up redirects for individual pages pointing visitors toward a relevant alternate page. This page should serve the same intent as the original, now-removed page.
For example, if you’ve just released a new model of a product, redirecting the page for the old model to the new serves ensures that users are landing on a page that serves the same purpose as the old.
Or if you’ve removed an old, out of date page and replaced it with a page that has the latest information on the subject.
As for the rest?
You’re better off creating an attractive, fun, light-hearted and (above all else) helpful 404 page and letting people see it.
Letting users and search engines land on a 404 page is better because it’s more candid and straightforward. It’s better to be upfront with a user and say a page isn’t available and disappoint them than it is to confuse them with a redirect.