Google's view of what's the main site in a network

How does Google determin the main site amongst multiple regional sites

[Whitey]

Hi Folks

We have several branded sites with different regional extensions [ e.g. .com ; co.uk ; etc ]

All sites are hosted in Canada on a single C block

Each site's content is differentiated for the local markets, but when we put the domain name into G minus the extension e.g. "widgetcompany" only one site appears, and it's not our preferred choice! It's one of the minor regional sites.

Now if I go to G and put "Yahoo" in the search it returns only Yahoo.COM. If I go to "G.co.uk" and do the same i get uk.Yahoo.com and Yahoo.co.uk - that's how i like it.

But how is Google determining our main site and why is our regional site not showing up on a regional search where it should belong?

[Quadrille]

Are the sites really unique - or clones with minor local variations?

Because the way you describe it, I suspect you are facing a duplicate content issue.

You'd probably do much better if you put all your content on .com, with 'local' folders if there really are differences. And 301 the old domains to that site.

That would much reduce your marketing duplication, remove all risk of content duplication, and concentrate your ranking.

If you have more than one language, keep thoe sites, as there is zero risk of duplication, little marketing gain (and a risk of customer confusion).

If it really is necessary to have seriously different local sites, then allow the sites to develop a genuine local character; that will make duplication issues much less likely, and local ranking more likely to be more appropriate.

But if it's really one site - then make it so.

[Whitey]

Although I was trying to see the reason for Google defining a main site, the issue of duplication had me thinking.

Do duplicate image alt tags cause a problem?

Maybe we overlooked this as they are all the same.

[Quadrille]

A bunch of identical ALT attributes is realy quite spammy; ALT has a specific function (nothing to do with SEO), and they should be used correctly - ie to indicate what the image is.

Better not to have them at all than have them all the same (ie 99% false!). They currently are no use to peole with visibility problems - and they are toxic to Google - which we know does read them.

[Whitey]

Sorry - i didn't make it clear. The alt tags are all different on one site. The problem is they are the same images with the same alt-tags on the other sites.

However, the image file names are all different between URL's.

[Quadrille]

Not so bad; the chances are that the ALTs are a pretty small proportion of the total page content - so provided the pages are unique, a line of code the same here and there should not be a problem in most cases.