So I was doing a bit of research today and came across a story that lead me to Allure Magazine and I saw a textbook example of how using nofollow the wrong way can work against you.
A lot of people take the “nofollow” tag quite literally thinking they can use it to herd the search engine spiders away from areas of their website they don’t want it to follow (which is actually understandable do to it’s name). However as was pointed out on a recent conference panel this is incorrect
SES Chicago: Search Engine Q&A On Links
nofollow is still used for discovery but it is not used for link popularity. The link is not treated as a vote. You cannot hurt yourself by using a nofollow link. So the link is still spidered.
So Allure magazine tried to keep Googlebot from filling out the form (which it can’t do yet) and ordering the magazine by surrounding it with “no-followed” links shown below.
However what they actually did was shoot themselves in the foot squandering the value of their own internal anchor text. for example look at the SERP for [subscribe allure magazine]. Looking at the results below
We can see they rank #5 for a phrase they should “own”. Instead they pay out commissions every month when they could fix the problem with about 15 minutes worth of work.
So what’s they take away here, understand how “no follow” works. Use a combination of robots.txt and meta tags “nofollow” and “noindex” to keep the spiders out. Lastly make sure keeping the page out of the index is really the right thing in the first place.
Related posts:
- After Taking 3500 Point Olympic Dive, Web Says: “Follow My Lead!” The economy is in a rut because of Big Risk...
- Wordpress SEO: How to Use RSS and Scrapers to Build Links When you run wordpress or any other blog for that...
- Thanks to this Months Sponsors – May 2009 I’d like to say thanks to the people who sponsored...
{ 21 comments }
What about Matt Cutts’ post on herding bots from July?
He stated:
“At a link level, you can add a nofollow tag on the granularity of individual links to prevent Googlebot from crawling individual links.”
Sorry, I forgot to include the direct link:
http://www.mattcutts.com/blog/bot-obedience-herding-googlebot/
“So Allure magazine tried to keep Googlebot from filling out the form”
Subscribe link points to:
http://www.condenastdirect.com/group/all_061113_split
They don’t need that page ranking in the SERPS.
“So what’s they take away here, understand how “no follow†works.”
What they’re doing is business as usual.
Nofollow will lower the chance of internal pages getting indexed by cutting off internal link juice flowing to those pages. META NOINDEX is obviously more reliable because it prevents Google from indexing a page even if other people’s sites point to a noindexed url. Whether Googlebot follows nofollow links or not is irrelevant as long as the target pages stay out of Google’s index.
How is keeping the “subscribe” anchor text and the value it could be passing to another page of value to them?
My guess is that the folks at Allure are trying to increase the link value of other links on the page by including nofollow on various links on the page that repeat across the site i.e. the header and footer rather than trying to stop Google from filling out a form.
But I agree with you in that they’re missing out on a fairly non-competitive phrase that they should own.
Anyone using nofollow for anything other than user-generated content runs the risk of shooting a foot.
I would suggest that most people would be better off never using ‘nofollow’:
- It doesn’t work for all spiders/search engines
- Even with Google, its use is poorly understood
- As the above discussion shows, one can hurt oneself
- If you don’t want a link to be followed, you probably shouldn’t have the link
- Some SEO professions believe that use of nofollow (or at least, excessive use) may be penalised by Google SERPs
- There are likely better methods in most cases (e.g. noindex)
Pages that have been blocked by robots.txt no longer appear in Google’s index, yet they still retain PageRank. Is this because the PageRank score we see is old data? I’m a major SEO n00b and even I have seen enough weird stuff that I don’t trust Google not to crawl whatever the hell it feels like.
I have seen Google crawl and index domains with no inbound links whatsoever. I though this was creepy until I read Google’s Webmaster Guidelines wherein it is written:
“It’s almost impossible to keep a web server secret by not publishing any links to it. As soon as someone follows a link from your “secret” server to another web server, your “secret” URL may appear in the referrer tag and can be stored and published by the other web server in its referrer log. So, if there’s a link to your “secret” web server or page on the web anywhere, it’s likely that Googlebot and other web crawlers will find it.”
We put an internal company forum on a URL with no inbound links (and few/no outbounds). Nobody used it, but a few dumbasses hotlinked some images – maybe that was the issue. After a while we went in there and found a bunch of classic “nice site buy cheep phentermine” automated spam. “How did they find us?” the boss asked me. I went to Google and typed site:www.domain.com and surprise surprise, there we were.
If I was a webmaster and I was seriously worried about bots crawling my forms I would consider putting in a captcha or doing some heavy php ninjitsu. nofollow or even robots.txt simply will not cut it.
Although, what’s interesting is that Google’s been working on ways to interpret the code and context of a form to determine what’s behind it. Refer to http://www.seobythesea.com/?p=337
Rand has a similar post regarding the copy that a searcher sees. Good call out, but I don’t agree with the following –
“We can see they rank #5 for a phrase they should “ownâ€. Instead they pay out commissions every month when they could fix the problem with about 15 minutes worth of work.”
They should definetly fix the error you have pointed out. However, they should definitely keep running their PPC ads, as long as they are hitting their ROI goals on their paid campaign. Just because you are ranking at the top of one result (paid or organic) does not mean you should exclude the other. Plenty of profit to be had in both.
oops, here is the link to Rand’s post http://www.seomoz.org/blogdetail.php?ID=1611
Robots.txt is so much easier and safer.
I never said get rid of the PPC, I just want a solid #1 PPC and #1 organic listing.
ahh, sorry I misinterpreted that
Nofollow isn’t as safe as it should or could be. Search engines will still check out why you are using the nofollow.
Nofollow is a leaky condom (pardon my words)
Pretty funny, huh
when company like this loses because of wrong know-how 
Although, typically magazines do not want their subscription links to be found because the offers change regularly.
the offer may change but wanting people to subscribe won’t
I think I figured it out.
I don’t think they are too concerned about Google filling out a form. Especially since the Googlebot would need a credit card to complete the form.
The links in the actual markup are fake links. There’s a script that replaces the links with tracking links. The dummy links are actually all broken links. They don’t seem to go anywhere.
If google were to index the dummy links, it would be a waste of PR because those links don’t go anywhere. See for yourself. Try following the links in the markup. They’re all broken! That would explain the nofollow.
Michael, your hypothesis seems to have some holes in it. Allure.com is still using rel=”nofollow” on all of their subscribe links, but they’ve moved up to #1 since your article was written.
Looks like some of the ranking algo has been tweaked
Then I guess this article just became irrelevant, then. There now seems to be no concrete evidence that nofollow actually has any effect on keywords.
Comments on this entry are closed.