Because nobody knows much about what goes on behind the scenes at Google, anything that sounds bad like a 404 error is automatically viewed as damaging.
Fortunately, for anyone tasked with trying to remove 404 errors, this is not necessarily the case and can be a waste of time and effort if those urls are no longer serving any kind of purpose.
When dealing with 404s, a lot depends on the history of the urls returning those errors. So next time you log into Google’s Search Console and see that alarming chart showing 404 errors, take a step back and assess what the causes of the errors are before acting on them.
404 errors are not in themselves a bad thing in all cases. 404s are simply created when a page url can no longer be found. This might happen, for example, on an e-commerce store where products are discontinued. The page is removed and a 404 is returned instead. Over time that page will eventually drop out of the search results altogether which can either be good or bad depending on what was featured on the page.
When 404s are bad
It would be wrong to say 404s don’t cause any problems. If any of the following cases apply then you will need to do something about the error(s).
• The page had lots of inbound links pointing to it
One of the most important ranking factors for a website are inbound links. Particular web pages may accumulate lots of inbound links over time if they serve useful information to their audience. Blog posts, homepages and useful information pages are the ones most likely to attract links. If any of these pages are returning a 404 then all the rankings for that page may be lost as well as a substantial amount of traffic to the site when this takes place.
• A new page has been created covering the same topic
If you have created a new page on the website covering the same topic when a previous one needed updating, then the old page should be permanently 301 redirected to the new one to preserve any benefits the old page might have had.
• Following a site migration
Site migrations and redesigns often open up the possibility of an increase in 404 errors. For example a page may cover the same topic with a slightly different theme meaning the old one is not redirected. This is why it is important to take a record of all the old urls of a website so that checks can be made following a site migration to ensure all urls are 301 redirected to their new homes.
When it’s Ok to Ignore 404s
As I mentioned earlier there will be plenty of cases where 404s are returned and this is ok if those urls have no inbound links and no regular traffic coming in. For example, you may have had a page dedicated to a one-off event but that event has now expired. It would be a bit daft to redirect that url somewhere else irrelevant just to avoid having a 404 error.
When it comes to 404s a degree of common sense is required when dealing with them.
404 errors will eventually fall away over time anyway when a page has expired, although they may still exist for a considerable amount of time. As long as you use Google search console and any of the other paid website SEO tools to identify all urls returning a 404, you should be able to handle any important errors that slip through the net.
Avoiding harmful 404s is all about preparation. Ensure all important urls are recorded prior to a site migration and analyse traffic and links to a page carefully before removing it from your website.
As John Mueller says in this video covering the subject, having 404s will not negatively impact the entire site.