No Follow to Homepage - Why are sites doing this?

[irldonalb]

I've recently followed matt cutts' instructions to highlight no-follow in Firefox [mattcutts.com].

Now when I browser the web, I see what other sites are doing.

I'm baffled by what I believe is misuse of the tag.

- A leading SEO blog / magazine has the tag on all links their homepage. Also within a page, they've the tag on some links to category pages. This ignores the "first link counts" logic spoken about previously here and at pubcon
- On a leading tech magazine (PR8), they've a no follow tag to their homepage from every page.
- Top UK portal / newspaper (PR7) has no follow tags to all their categories including homepage

Do they know something I don’t know?

Thanks
Donal

[Robert Charlton]

Back to the OP's rel="nofollow" question...

Now when I browser the web, I see what other sites are doing.

I'm baffled by what I believe is misuse of the tag.

When I browsed the web keeping this question in mind, I saw a variety of nofollow usage, much of which might lead to unforeseen problems. In general, the rel="nofollow" seems to be the device that is often being called on to fix either structural issues or speculative issues.

I'm seeing some sites that are nofollowing links to the current page (ie, links to home from home, or to any page from itself).

Also frequently nofollowed are links from dupe nav menus (eg, from a top or bottom copy of the same menu)... And also nofollowed are Read More type links, etc... ie, control of anchor text.

Some efforts apparently want to de-emphasize the ranking effect of the home page or of higher level category pages. These frequently contain many goof-ups. One site I encountered appears to have nofollowed all links that had the word "home" as the anchor, but had left no useful alternative... and there was no alt anchor in the graphic logo link, which was the only dofollow link I could find back to home.

Another site didn't link to back to home at all, but did link to all of its main nav categories via the left hand nav menu... so these weren't exactly being siloed. This site was particularly confused/confusing, because its template had an individualized heading on each page whose text identified each current page... but each of these current-page headings then linked to home, and all the links to home, including these heading links, were then nofollowed. ;)

newspaper (PR7) has no follow tags to all their categories including homepage

After seeing some of the above examples and others, I'm thinking this may be an effort to keep category pages from ranking above a specific story page. Not sure how link juice actually would get down to lower level pages. Many of the nofollow efforts I see are trying to control which internal duplicate content ranks. I feel a lot of PR and link juice is likely being lost in the latter process... of repressing internal dupes.

In general, you don't want to channel either too much or too little PR to deeper pages... either is wasteful. But excessive use of rel="nofollow" can waste it for the whole site.

[TheMadScientist]

One of the things about this whole topic is and why I think there may be different results from testing is as a search engine you can write so many different conditions into the if()s it's ridiculous, and you don't really have to treat any two situations the same way and how you vary the treatment slightly can be based on anything you have access to...

if (there's only one link to the page, treat the nofollow as ...)
if (there are 3 links to the page from the page && one is nofollowed ...)
if (there are 3 links to the page && 2 are nofollowed ...)
if (there are 2 links to the page && 1 is nofollowed && the link text in the other link is 'click here' || (or) the link text in the other link is 'see more' || (or) the other link text is 'home' || (or) the other link text is 'domainname.com' ...)
if (the count of the links to the page is > 1 && the count of the nofollowed links to the page / the count of the links to the page is <= .5 ...)
if (the total number of links to the page in the DB is > a threshold number && the total number of the links nofollowed / total links on the page to the page <= .3 ...)

I could seriously do this for hours and probably 1000+ lines of code (just to figure out how to handle a stinking nofollow link(s) on a page) if I started getting into PR and other factors, and IMO from having written code, to think anything Google has in it's algo is not 'fuzzy logic' that can be adjusted by many factors is probably lack of coding knowledge, because it doesn't take a team of programmers to write a statement to say 'nofollow=action' and apply it to every single case, but to be able to 'adjust, weight and vary' factors until you get a better set of results does...

### # ###

Anyway, I think sometimes we (I do this too) get a result and think 'that's the final answer' and don't really stop and think how complicated an algo with 200 factors is, and if you think it's just straight through factor=action, you're probably really mistaken... I know MC gets garbage for being noncommittal sometimes, but what people don't realize is how many factors or 'options' go into something, so 'uh, er, usually you want to... blah' or 'in most situations... ' is really the right answer.

If you watch the video in linked in the other Internal nofollow thread he doesn't say to 'not ever use nofollow internally' as some are trying to say, if you listen to the exact wording he uses he says 'usually' or 'in most situations' or 'relatively, no' because it CAN often times do more harm than good, but he doesn't say 'always' or 'don't ever for any reason' as some people are trying to say.

[edited by: TheMadScientist at 7:01 pm (utc) on Jul 6, 2010]

[TheMadScientist]

Here's the link to the other thread...

PR Sculpting Doesn't Work and Internal NoFollow Can Harm Your Site [webmasterworld.com]

Listen carefully before you think the 'absolute' posted in the thread as 'the answer' is actually what he says, because he really does not give an absolute in the video. He says things like 'usually' and 'in most cases'.

[johnser]

I think…..
“If one occurence of a linked URL is "not trusted" why would a second one be trusted, anyway?”

Also….

“Someone mentioned before that no-follow tags on authority sites can have a different behaviour. It’s possible that the no-follow tags on the sites I mentioned above are being ignored”

AFAIK, Google counts no-follows from authority site like Wiki – when the link has been in place for a good while – which makes perfect sense.
If only some outbound links on an auth site are no-follow, then it’s likely that G would ignore them

If there are internal link no-follows, G probably devalues all internal PR to that page. Which would make sense in the case of T&C’s or fluff pages.
Sites are storing up problems for themselves doing this with the home page link I would imagine.

If it makes logical sense to do something, do it. If not, don’t.

[tedster]

AFAIK, Google counts no-follows from authority site like Wiki – when the link has been in place for a good while

Not as far as I know. Here's a direct "no" answer on that question from a Matt Cutts video. [youtube.com...]

Have you tested and discovered otherwise?

[irldonalb]

It's worth pointing out that Google follow links for different reasons. I think Google "follow" no-follow links for discovery purposes. However that link mightn't count in its PageRank algorithm.

Hence why I'm suspicious when people are quick to publish results of their tests.

Hypothetical test:

Page A links to Page B with a no-follow tag. This is only place Page B is linked to. Page B gets cached and ranks for some meaningless long term phrase. Therefore Google must ignore no-follow tags.

That's not a real world test. It's results are probably completely different to when Page B has 100 inbound links; one of which is a no follow tag. And the phrase in question isn't some sudo phrase no one has searched for previously.

[mhansen]

The fact pagerank and nofollow even exist is what's creating these problems.

Couldn't agree more! Could all of this nofollow FUD stuff be GooG's way of skunking out over optimization?