How Important Is Redirecting Old Broken Links?

[Planet13]

Hi there, Everyone:

Webmaster tools is showing me lots of old broken links. How important is it to fix everyone of them?

many of them were the cause of my content management system previously allowing multiple versions of a URL to resolve to the same page. so for example:

mydomain.com//Widgets/Product1.html
mydomain.com//FooBars/Product1.html
mydomain.com//Gizmos/Product1.html

All resolved to the same product page, namely, Product1.html

do I really need to go and correct my htaccess file with redirects to the current page? I don't see any EXTERNAL sites linking into those pages. All the links to those pages listed by GWT were said to be internal from my own site. so I don't think I would be squandering any page rank by NOT redirecting them, would I?

[Sgt_Kickaxe]

If no external sites are linking to the pages and you have no internal links pointing to them it's not going to be of much benefit to redirect them at this point. I say that under the assumption that those old urls are returning a 404 or 410 header response, if they aren't they need to be.

You could make an argument that they should be redirected since they existed and now don't but in reality they are gone and won't be returning so 404/410 are the expected results. The argument against is having 301 redirects leading FROM a 404/410 page might look like spam. If you want to redirect them they should probably still be there to avoid the appearance of spam. People redirect empty domains for ranking purposes all the time so it's not behavior Google doesn't look for.

I don't mean to scare you, Google has history and knows they once existed no doubt but it's best to stay safe.

In the future you can leave them up if they make sense for visitors and use a canonical tag but ideally it's best not to have multiple copies of one page.

[Planet13]

Thank you, Sgt_Kickaxe:

i will run xenu tonight to make sure that there aren't any internal links pointing to them. I would rather play it safe and just let them return 404 pages.

[lucy24]

I say that under the assumption that those old urls are returning a 404 or 410 header response

A 410 doesn't happen automatically. You have to give explicit directions. The redirect is not coming from the 404; once anyone (robot or human) hits a redirect, they will never know whether the address they asked for exists or not. So far, google has not figured out how to override htaccess or config files.

[Planet13]

Hi ya, Lucy24:

Thanks for the note.

I will check them with fetch as googlebot tomorrow.

But I thought that since they were showing up as crawl errors in GWT, that pretty much meant that google is seeing a 404 (or 410) header.

Again, you know what they about assuming, so I will double check in the 'morn.