Google Webmaster Tools Link Data Is Only A Sample

Jun 30, 2011 • 8:35 am | comments (7) by twitter Google+ | Filed Under Google Search Engine Optimization
 

Google WebmasterGoogle Webmaster Tools has a very powerful tool for SEOs, it is the links to your site report.

Ever since launching that report, Googlers and SEOs have been encouraging people to stop using the link command and use this tool. In fact, on the Google help page it said, "using the link: operator to find a sampling of links to any site." They then suggest you use the link tool within webmaster tools to get "a much larger sampling of links."

So when I spotted a post at Google Webmaster Help, I noticed a person complaining that the tool is only showing 1% of their real links. In response to that, Googler Pierre Far replied:

Keep in mind also that Webmaster Tools show a sample of links, but isn't guaranteed to show 100%. So some links may exist to either URL that aren't necessarily displayed in Webmaster Tools.

Pierre was not the first Googler to say this about webmaster tools, Susan Moskwa said it in March 2010, but I missed it then.

I am surprised that would be the response for a tool that shows a pretty accurate number of links. Google has been pushing people to stop using the link command and use this tool. If it is just a "sample of links" and not that accurate, what is the point?

This is a bit upsetting to me. Not that the link report might be not a 100% accurate but that Googlers are using the same excuse they used with the link command here with the link tool within Webmaster Tools.

Forum discussion at Google Webmaster Help.

Previous story: Google Calendar Redesign Live
 

Comments:

Wouter Blom - Bureau Stramark

06/30/2011 01:17 pm

lots of other data in webmastertools is also not correct, or only for pages that have been online ages ago. I got a broken link report from a sitemap.gz which was online early 2009....

Michael Martinez

06/30/2011 05:19 pm

They have always said that it would not show 100% of links.

Doc Sheldon

06/30/2011 05:44 pm

I share your frustration, Barry, although I don't think it's realistic to expect a high degree of accuracy from a free tool, even one offered by Google. Since we seem to be in the throes of a major Google facelift, maybe some enhanced definition is in the offing? Dare we hope? And maybe they'll even give us back the Wonder Wheel! ;)

Barry Schwartz

06/30/2011 05:45 pm

Yea, but I don't expect Google to reply to the webmaster who says it shows 1% of their links as "this is sample data." Too different things.

Doc Sheldon

06/30/2011 05:55 pm

True. I've never seen a case like that, but I suppose if they count their links in the hundreds of thousands, that might well be the case. What I find really puzzling is the variation in the numbers one can get between those two methods. There seems to be no rhyme or reason to it.

Dharne

07/04/2011 06:31 am

Subset of data may be better than wrong data, but we won't know if it is the wrong subset either.  Well, it is a free tool after all.

Bruno

09/02/2011 02:57 pm

Sample is an euphemism.

blog comments powered by Disqus