They are called sitelinks - to get them you seem to need to have a website that is particularly hierarchical and strong traffic - but that's just my hunch.
Just realised I didn't actually answer your question fully - at the moment there is nothing you can personally do to get sitelinks - they are algorithmically generated.
If anybody knows of any low traffic 'ma and pa' style sites with sitelinks I'd love to know about them - it would be good to analyse such a site and try and identify what the alg might be looking for.
> Just realised I didn't actually answer your question fully - at the > moment there is nothing you can personally do to get sitelinks - they > are algorithmically generated.
> If anybody knows of any low traffic 'ma and pa' style sites with > sitelinks I'd love to know about them - it would be good to analyse > such a site and try and identify what the alg might be looking for.
> Cheers,
> doc
> On Aug 26, 6:56 pm, Alaa Maanawi wrote:
> > Dear all any body know how to set the search results of website aswww.cnn.comtoshow the most important links on the website
> > go to Google search box : search for 'cnn.com' results come this way:
Something else that I think it needs is to be a really strong #1 for the search term, it almost has to be a navigational query (ie you're really just trying to go to the site in question, nothing else). If you try "cnn" you'll always end up going to www.cnn.com. The sitelinks give you a head start into the part of the site that you might be interested in going to, since Google already knows that you want to go there. You search for "cnn" and it asks you "which part of cnn?".
I see that on my old site, [gsitecrawler] brings up a page from the site (I guess that's a sign that it isn't the strongest result by far), [gsitecrawler.com] brings up the sitelinks.
My guess is that it does something like this:
- sites that have "enough value" are crawled "better" - if that "better" crawl (+ perhaps the value of the individual pages?) can bring out a strong structure (think site-wide links, consistent titles / anchor texts), it can be used for sitelinks - if a navigational query is made, where the site is in a very strong position, it will display those sitelinks
If you look the the GSiteCrawler sitelinks and site, you will see that some of the text used for sitelinks comes from the anchor text ("Download", "Features", "Tour") but some also comes from the page title ("FAQs and Documentation"). Maybe "FAQ" is just too short? At any rate, I imagine both have to match, more or less.
Something neat I just noticed, if you do [cnn headlines], it has a sitelink "sports" which points at www.cnn.com/si/?cnn=yes -- but that 302 redirects to http://sportsillustrated.cnn.com/?cnn=yes ... so I guess the links count more than the content for sitelinks? or is it another "the shorter url wins"?
I've also seen subdomains listed in sitelinks, which I found a bit interesting considering the 'subdomains are treated as completely separate sites' mantra.
re: your gsitecrawler search - I get the sitelinks even when I type gsitecrawler only.
> Something else that I think it needs is to be a really strong #1 for > the search term, it almost has to be a navigational query (ie you're > really just trying to go to the site in question, nothing else). If > you try "cnn" you'll always end up going towww.cnn.com. The sitelinks > give you a head start into the part of the site that you might be > interested in going to, since Google already knows that you want to go > there. You search for "cnn" and it asks you "which part of cnn?".
> I see that on my old site, [gsitecrawler] brings up a page from the > site (I guess that's a sign that it isn't the strongest result by > far), [gsitecrawler.com] brings up the sitelinks.
> My guess is that it does something like this:
> - sites that have "enough value" are crawled "better" > - if that "better" crawl (+ perhaps the value of the individual > pages?) can bring out a strong structure (think site-wide links, > consistent titles / anchor texts), it can be used for sitelinks > - if a navigational query is made, where the site is in a very strong > position, it will display those sitelinks
> If you look the the GSiteCrawler sitelinks and site, you will see that > some of the text used for sitelinks comes from the anchor text > ("Download", "Features", "Tour") but some also comes from the page > title ("FAQs and Documentation"). Maybe "FAQ" is just too short? At > any rate, I imagine both have to match, more or less.
> Something neat I just noticed, if you do [cnn headlines], it has a > sitelink "sports" which points atwww.cnn.com/si/?cnn=yes-- but that > 302 redirects tohttp://sportsillustrated.cnn.com/?cnn=yes... so I > guess the links count more than the content for sitelinks? or is it > another "the shorter url wins"?
Another little interesting aside - wondering 'where' Google gets the headings for the sitelinks -
In your case John, it's obvious that google has used the nav bar at the top as the 'most important links' - but where it actually gets the anchor text seems to differ...
1. Download - comes from anchor text in header bar, as the page title is 'Downloads' - plural 2. FAQ's and Documentation - comes from the page title, as the anchor text is simply 'FAQ' 3. Features - both anchor text and title are the same. 4. Tour - comes from the anchor text in the nav bar, as the page title is 'Tours of a Sitemap Generator'
It's weird that google seems to 'pick and choose' like that.
> I've also seen subdomains listed in sitelinks, which I found a bit > interesting considering the 'subdomains are treated as completely > separate sites' mantra.
> re: your gsitecrawler search - I get the sitelinks even when I type > gsitecrawler only.
> M
> On Aug 26, 7:36 pm, JohnMu wrote:
> > Something else that I think it needs is to be a really strong #1 for > > the search term, it almost has to be a navigational query (ie you're > > really just trying to go to the site in question, nothing else). If > > you try "cnn" you'll always end up going towww.cnn.com. The sitelinks > > give you a head start into the part of the site that you might be > > interested in going to, since Google already knows that you want to go > > there. You search for "cnn" and it asks you "which part of cnn?".
> > I see that on my old site, [gsitecrawler] brings up a page from the > > site (I guess that's a sign that it isn't the strongest result by > > far), [gsitecrawler.com] brings up the sitelinks.
> > My guess is that it does something like this:
> > - sites that have "enough value" are crawled "better" > > - if that "better" crawl (+ perhaps the value of the individual > > pages?) can bring out a strong structure (think site-wide links, > > consistent titles / anchor texts), it can be used for sitelinks > > - if a navigational query is made, where the site is in a very strong > > position, it will display those sitelinks
> > If you look the the GSiteCrawler sitelinks and site, you will see that > > some of the text used for sitelinks comes from the anchor text > > ("Download", "Features", "Tour") but some also comes from the page > > title ("FAQs and Documentation"). Maybe "FAQ" is just too short? At > > any rate, I imagine both have to match, more or less.
> > Something neat I just noticed, if you do [cnn headlines], it has a > > sitelink "sports" which points atwww.cnn.com/si/?cnn=yes--but that > > 302 redirects tohttp://sportsillustrated.cnn.com/?cnn=yes... so I > > guess the links count more than the content for sitelinks? or is it > > another "the shorter url wins"?
so its only on hand of google i think of how the links used for the term on the anchor text and the time spent on the page its reflect how important its back google list this link in the search result of the domain! Thanks guys for answers
> Another little interesting aside - wondering 'where' Google gets the > headings for the sitelinks -
> In your case John, it's obvious that google has used the nav bar at > the top as the 'most important links' - but where it actually gets the > anchor text seems to differ...
> 1. Download - comes from anchor text in header bar, as the page title > is 'Downloads' - plural > 2. FAQ's and Documentation - comes from the page title, as the anchor > text is simply 'FAQ' > 3. Features - both anchor text and title are the same. > 4. Tour - comes from the anchor text in the nav bar, as the page title > is 'Tours of a Sitemap Generator'
> It's weird that google seems to 'pick and choose' like that.
> M
> On Aug 26, 7:45 pm, dockarl wrote:
> > I've also seen subdomains listed in sitelinks, which I found a bit > > interesting considering the 'subdomains are treated as completely > > separate sites' mantra.
> > re: your gsitecrawler search - I get the sitelinks even when I type > > gsitecrawler only.
> > M
> > On Aug 26, 7:36 pm, JohnMu wrote:
> > > Something else that I think it needs is to be a really strong #1 for > > > the search term, it almost has to be a navigational query (ie you're > > > really just trying to go to the site in question, nothing else). If > > > you try "cnn" you'll always end up going towww.cnn.com. The sitelinks > > > give you a head start into the part of the site that you might be > > > interested in going to, since Google already knows that you want to go > > > there. You search for "cnn" and it asks you "which part of cnn?".
> > > I see that on my old site, [gsitecrawler] brings up a page from the > > > site (I guess that's a sign that it isn't the strongest result by > > > far), [gsitecrawler.com] brings up the sitelinks.
> > > My guess is that it does something like this:
> > > - sites that have "enough value" are crawled "better" > > > - if that "better" crawl (+ perhaps the value of the individual > > > pages?) can bring out a strong structure (think site-wide links, > > > consistent titles / anchor texts), it can be used for sitelinks > > > - if a navigational query is made, where the site is in a very strong > > > position, it will display those sitelinks
> > > If you look the the GSiteCrawler sitelinks and site, you will see that > > > some of the text used for sitelinks comes from the anchor text > > > ("Download", "Features", "Tour") but some also comes from the page > > > title ("FAQs and Documentation"). Maybe "FAQ" is just too short? At > > > any rate, I imagine both have to match, more or less.
> > > Something neat I just noticed, if you do [cnn headlines], it has a > > > sitelink "sports" which points atwww.cnn.com/si/?cnn=yes--butthat > > > 302 redirects tohttp://sportsillustrated.cnn.com/?cnn=yes... so I > > > guess the links count more than the content for sitelinks? or is it > > > another "the shorter url wins"?
> so its only on hand of google > i think of how the links used for the term on the anchor text and the > time spent on the page its reflect how important its back google list > this link in the search result of the domain! > Thanks guys for answers
> On Aug 26, 1:54 pm, dockarl wrote:
> > Another little interesting aside - wondering 'where' Google gets the > > headings for the sitelinks -
> > In your case John, it's obvious that google has used the nav bar at > > the top as the 'most important links' - but where it actually gets the > > anchor text seems to differ...
> > 1. Download - comes from anchor text in header bar, as the page title > > is 'Downloads' - plural > > 2. FAQ's and Documentation - comes from the page title, as the anchor > > text is simply 'FAQ' > > 3. Features - both anchor text and title are the same. > > 4. Tour - comes from the anchor text in the nav bar, as the page title > > is 'Tours of a Sitemap Generator'
> > It's weird that google seems to 'pick and choose' like that.
> > M
> > On Aug 26, 7:45 pm, dockarl wrote:
> > > I've also seen subdomains listed in sitelinks, which I found a bit > > > interesting considering the 'subdomains are treated as completely > > > separate sites' mantra.
> > > re: your gsitecrawler search - I get the sitelinks even when I type > > > gsitecrawler only.
> > > M
> > > On Aug 26, 7:36 pm, JohnMu wrote:
> > > > Something else that I think it needs is to be a really strong #1 for > > > > the search term, it almost has to be a navigational query (ie you're > > > > really just trying to go to the site in question, nothing else). If > > > > you try "cnn" you'll always end up going towww.cnn.com. The sitelinks > > > > give you a head start into the part of the site that you might be > > > > interested in going to, since Google already knows that you want to go > > > > there. You search for "cnn" and it asks you "which part of cnn?".
> > > > I see that on my old site, [gsitecrawler] brings up a page from the > > > > site (I guess that's a sign that it isn't the strongest result by > > > > far), [gsitecrawler.com] brings up the sitelinks.
> > > > My guess is that it does something like this:
> > > > - sites that have "enough value" are crawled "better" > > > > - if that "better" crawl (+ perhaps the value of the individual > > > > pages?) can bring out a strong structure (think site-wide links, > > > > consistent titles / anchor texts), it can be used for sitelinks > > > > - if a navigational query is made, where the site is in a very strong > > > > position, it will display those sitelinks
> > > > If you look the the GSiteCrawler sitelinks and site, you will see that > > > > some of the text used for sitelinks comes from the anchor text > > > > ("Download", "Features", "Tour") but some also comes from the page > > > > title ("FAQs and Documentation"). Maybe "FAQ" is just too short? At > > > > any rate, I imagine both have to match, more or less.
> > > > Something neat I just noticed, if you do [cnn headlines], it has a > > > > sitelink "sports" which points atwww.cnn.com/si/?cnn=yes--butthat > > > > 302 redirects tohttp://sportsillustrated.cnn.com/?cnn=yes... so I > > > > guess the links count more than the content for sitelinks? or is it > > > > another "the shorter url wins"?
