Webmaster Tools finding phantom 404s?

[Switch_Digital]

We recently (three months now!) switched over a site from .co.uk to .com and all old urls are re-directing to the new site.

However, Google Webmaster tools is flagging up hundreds of 404s from the old site and yet doesn't report where the links were found, i.e. in the 'Linked From' tab there is no data and the old links are not in the sitemap.

SEOmoz crawls do not report any 404s.

Any ideas?

[Switch_Digital]

Okay, thanks, and you are quite right, a whole morning is more than enough energy... gotta love SEO work!!!!

[Nobody1560986989723]

OK, well if it truly doesn't make sense (does sound odd, and it does seem like you've done the redirects fine) and three months is more than long enough for GWT to have caught up I'd take the above approach and periodically download the 404 list and seeing if there are any additions, as well as seeing if maybe Bing Webmaster Tools agrees with GWT.

If everything is redirecting fine, then I'd be inclined to just disregard it for the time being and focus my energies elsewhere. Good luck with it!

[Nobody1560986989723]

Maybe temporarily stop the 301 on the old site. Re-run your crawl reports and see if there were any 404s in existence on the old site that you hadn't previously thought. Plug the links and then reinstate the 301?

Either that or, if you're sure there's no problem, download the phantom 404s to CSV and then only take note of additions to that list in future?

[Switch_Digital]

That's the strange thing, all urls on the .co.uk are 301'ing to the new site.

[Nobody1560986989723]

If you can't tell where the links are coming from then the next bet is as Ben said is to identify the URLs which are being linked to and 301 them to a page which is closely related (or the home page otherwise), that way you don't lose any potential visitors to your site via those links.

[Switch_Digital]

Webmaster is picking up 494 pages returning a 404 and yet Roger Bot is finding none, so I can't even identify where the links are coming from. This is the part that is confusing me.

If I try and access any of the pages reported in the list,  I am simply re-directed to the new site, as we are re-directing all .co.uk pages to the .com

[BenRWoodard]

I've had similar experiences after migrating sites over to a new CMS.  The links are being picked up from somewhere so the best thing to do is create a 301 redirect if there is a valuable similar page to the 404.  The only other option is to lose possible link juice from where ever the link is coming from.  It's always better to have a pure link (no 301) but if you can't contact the original link then there is nothing you can do.

I'm curious as to where the links come from as well.  Maybe someone will be able to speak to that.