Quick comment on nofollow

The rel=”nofollow” attribute is an easy way for a website to tell search engines that the website can’t or doesn’t want to vouch for a link. The best-known use for nofollow is blog comment spam, but the mechanism is completely general. Nofollow is recommended anywhere that links can’t be vouched for. If your logs analysis program shows referrers as hyperlinks, I’d recommend using nofollow on those links. If you have a wiki that anyone on the web can edit, I’d recommend nofollow on those links until you can find a way to trust those links. In general, if you have an application that allows others to add links, web spammers will eventually find your pages and start annoying you.

Let me give an example to illustrate. There’s a domain that runs an oompa loompa dating service. Oompa-Loompas are the small folks from the original Charlie and the Chocolate Factory. I think that the dating service is just a gag; it’s a fun way that people can play around and pretend to be oompa loompas. It used to have real people leaving messages for each other. But it also lets you add a link to a webpage, so this fun service has been inundated with people trying to get links. In the picture below, notice that every comment is pretty meaningless: “Good content and very informativity! Thanks!” and “Your website has been very helpfull to me!!”. And if you mouse over the little home page icon, you see why; I’ve highlighted one below:

Oompa Loompa Dating!

The fact that webspammers will find and attack a one-off application is very telling. It shows that if you run a site that lets anyone add a regular link, webspammers will eventually find your site and spam it as well.

I’d be the first to say that nofollow isn’t perfect. For example, plenty of people will set their bots loose, and those bots will spam for links without checking if a particular page has nofollow. But the people that write the bots also aren’t dumb. If it doesn’t add any benefit to spam a particular software package, a smart spammer will avoid wasting the time/effort on that software.

If you run a well-known website or software package, webspam is more of an issue for you. Someone recently pointed me to this wikipedia thread, where someone asked if Google was in favor of enabling nofollow on wikis, so I wanted to give a quick reply: I do think it’s a good idea. For example, I’ve talked to a couple SEOs recently who said that they have a full-time person on their staff dedicated to scamming links from Wikipedia and wikis.

In an ideal world, nofollow would only be for untrusted links. Let’s take the example of a forum that wants to avoid linking to spam, but the same advice applies to wikis or any other web software. If an off-domain link is made by an anonymous or unauthenticated user, I’d use nofollow on that link. Once a user has done a certain number of posts/edits, or has been around for long enough to build up trust, then those nofollows could be removed and the links could be trusted. Anytime you have a user that you’d trust, there’s no need to use nofollow links.

122 Responses to Quick comment on nofollow (Leave a comment)

  1. Matt, good article. I noticed this no-follow tag on your site last week after a post and thought it was cool. You’re also implementing CAPTCHA on this site so bots shouldn’t be able to SPAM it. Is that not working 100%?

  2. Hey Matt,

    This is the perfect segué to a question I’ve always had but never really bothered to ask.

    Since there are blogs, such as yours, that have regulars that do post and that have at the very least established themselves as people who aren’t here to spam, would the same logic not apply to blogs as well? i.e. remove the nofollow attribute on those links provided by helpful or at least non-hurtful types?

  3. Alex Duffield

    Matt, I have been having problems competing with sites that use Linkfarming techniques, and have reported these to Google when I find them, but Google doesn’t ever do any thing about them.

    I think the only way to stop people from using these methods is to penalize them when they get caught.

    My problem is now I am getting pressure from my clients to use the same techniques, and as much as I say it’s a bad idea, the fact is it works, providing the spammer with a #1 result in many of my clients market.

    Id be more than happy to provide you with details on the spammer if you would like!!!

  4. kapa

    Interesting idea,

    With all due respect, it seems to me that you are suggesting that all the non-commercial websites out there that get a link from Wikipedia (because for me this is what this post is all about) and a little PageRank should not anymore do that.
    So you are actually sending all these folk to spammers and evil SEO’s because lets face it, if you start a small non-commercial website noone wants to link to you Matt and despite the Google promises no one will ever find you either…not even the robots…

    Please correct me if I am wrong.

  5. I’m happy with how my captcha reduces spam, plus moderating comments also cuts things down. The main thing I wanted to do was provide an opinion for the wiki folks on the record.

    Adam, I use a stock WordPress install, but I’d certainly support the notion of registered/trusted users eventually gaining the ability to garner links. That said, I’m too busy to do it right now. :) The best bet would be going to the source: asking blog software or forum makers to provide those levels of granularity.

  6. Matt, it’s not all about Google. Spammers who continue to spam NOFOLLOW-linking sites do so because they know they can still scrape traffic from curioous click-throughs.

    NOFOLLOW continues to be a very bad idea, and one which I will continue to oppose. While blog comment spam is one aspect of the issue, the real problem is Google Adwords and Yahoo! Internet Marketing.

    Your companies pay the spammers and keep them in business. You’re trying to get the Web community to do your policing for you.

    You’re not paying me enough to implement something as ineffective as NOFOLLOW.

  7. Stephen

    Does Google think that everyone has no-follow on there homepages at the moment as Googlebot is only indexing homepages for some sites ? – Looking at the feedback at google groups – I still feel that 90-95% have duplicate homepages non-www/www etc…..intresting that although some sites have been reduced to homepage only – Google is listing multiple versions still eg domain.com, http://www.domain.com and http://www.domain.com/index.html etc

    Talking about Oompa Loompas (and to make this more on topic ;)) – in the new Charlie and the Chocalate factory I dont think it really worked cloning the one actor.

    Hmmz – why were you on a Oompa Loompa dating site one wonders.

  8. How about these uses:

    #1 – I have a blog that has a write-up on the latest “garden rakes”, since I do not sell rakes and am a kind of “consumer reports” I link out to Amazon.com but in doing so I am casting a vote for them. Long story short, the Amazon page for “garden rakes” appears high in the search because everyone links to the Amazon’s, Wiki’s and About.com’s of the world by habit right? I do not want to make them bigger, nofollow?

    #2 – If a site is a blog and the links that appear in the sidebar are “sitewide links” (meaning they appear on all pages), how about putting a nofollow on the useless “contact” page?

    #3 – I find a new template that someone made for my WordPress blog but a URL in the footer (to the guys site) also appears “sitewide”, how about a nofollow there?

    #4 – The same as #3 but the link is for some sort of stat program owned by a group to get PR all in the name of performancing your blog? nofollow?

    #5 – I have 5 blogs, I would like to link them all to each other (for my visitors) in the footer but feel it might be a little spammy, nofollow?

    You want more, just ask, this stuff is in great need of clarification.

    Thanks Matt.

  9. Matt, I’ve been looking for an opportunity to ask a related question around giving the SE’s help on what we want to count and not count.

    Say I have a community website or message board and every user has their own profile page. Not only do I not want unproven users to link out, but I’d like to keep their profile from even getting indexed until they earn it.

    So, for the new members, they’re geting a NOINDEX, NOFOLLOW robots meta tag. Is this an effective way to *temporarily* keep a page out of Google? Will gbot check again frequently to see if the NOINDEX tag has been removed? Or does it remove the page from the crawl list for a long time?

  10. I notice that you don’t mention paid links here, although you have before. Have you given up on having the webmaster community let you know that a link is an advertisement, and if so, are you making any headway in determining the nature of such links?

    As far as I’m concerned, that’s my only real problem with the nofollow idea. For blog comments, it makes perfect sense, especially in providing a disincentive to spammers, but advertisers are going to want the algorithmic benefit of the links they buy on top of any traffic they might receive. I’m afraid it’s just naive for the search engines to hope for site owners to voluntarily give up one of the selling points for advertising on their pages.

    And by the way, Yahoo has been counting my comments on your blog as backlinks to my site. This is apparently not a perfect system.

  11. kapa

    I totally agree with Bob but I am also curious and have a question for Matt:

    - Is organic search an academic citation tool or a commercial reality?

    Just trying to understand…

  12. Bloop

    norel is inefficient and I still cant belive google decided to roll with it.
    I cant see a reason why anyone would not put nofollow on ALL their outgoing links that do not belong to them, I mean not doing so would just cause PR leakage on your site + get you ranked lower.

  13. Matt, it is a very interesting article, but I think that Google’s own methods for determining the value of the site is at cause for this stuff.

    The only real way to get a site to have better SERPs after you have done reasonable white hat SEO on the site is to get “votes”. If you have a site that few people would vote for (example being a commercial camera sales page, most people would rather send those hits to an affiliate program), then you must resort to somewhat grey hat techniques to find those links and votes.

    By it’s very nature, the Google algo causes people to go out and try to abuse the system… and more often than not it works.

    With links and anchor text appearing to be a major part of a site’s SERP, it is no wonder people are willing to go to extremes to take advantage of the system. Maybe rather than nofollow links the algo should get a little smarter?

  14. Harith

    Bob Gladstein

    “For blog comments, it makes perfect sense, especially in providing a disincentive to spammers, but advertisers are going to want the algorithmic benefit of the links they buy on top of any traffic they might receive.”

    My experience tells me that advertisers whom most interested in traffic (maybe targeted) aren’t the main problem. Those folks doen’t mind the rel=nofollow , and mostly they even don’t know what it is.

    As to “algorithmic benefit of the links”. The problem, IMO, is the owners of sites of PR10, PR9 and PR8 who are selling textlinks mostly for one reason only; to boost PageRank. Clients demand such PR boost of the SEO specialists, because their compititors have already achieved higher PR by purchasing textlink etc…
    That put the whitehat SEO specialist under huge pressure. If you don’t do it, somebody else will.

    I’m sure all decent white SEOs know what I’m talking about here.

    I guess we need an open discussion among whitehat SEOs and Google, about how to handle the current situation.

  15. I believe 1-5 (listed above) benefit webmasters, anyone?

    I listed my site in Wiki because it is on a highly relevant page and the editors have not deleted it (like they did to several others who were in fact spammers). Wiki is self policing so yeah, I wouldn’t worry about that as much, but I think Matt might have been focusing on how spam can turn a good site into a bad neighborhood… yes/no?

  16. Chris Purcell

    Here’s a thought. How about Google dedicates a single person to looking for wikis et cetera that are being spammed and, to use common web terminology, b**ch-slapping the perpetrators? The wiki community would be more than willing to help set up a system to make it easy for you to find such spam attacks. Take chongqed.org, for instance: they maintain a database of exactly the information you need: namely, URLs that are being spammed, and links to diffs proving those URLs are being used. Heck, you could probably offer the chongqed guys a freelance contract to do it for you.

    If SEOs discover their spamming tactics are actually getting their sites a PageRank of zero, they will probably stop funding spam.

    As with your other heavy-handed techniques, you can also offer a “get my site back on Google” setup to allow the evil ones to repent.

  17. I’m always a lil’ afraid that one day a search engine comes along that uses “nofollow” as a negative vote, instead of no vote. ‘Course that’s not supposed to happen and I’m sure Google and others won’t do that. Other than that I like and use the “link condom.” I even use it on some of my internal links, e.g. on a link which I’m using repeatedly on the same page (not sure if that is perfectly good use).

    One thing though, when you say it’s good for wikis to use nofollow, I disagree, or rather, I agree on your “until you can find a way to trust those links”. A Wiki is self-correcting so I think it’s a much better plan to automatize this and use “nofollow” for the first week or so of an edit (I was pondering doing the same for my blog forum). If the link is stable and not edited out of the page, then it will get its linkjuice after a week when the nofollow is removed.

    Applied to Wikipedia, where this discussion also erupted — AFAIK non-English versions use nofollow, whereas the English one doesn’t — I would say this approach is as good as the rest of the system, i.e. something like “works 90% well, but trying to improve on the remaining 10% will break the whole system.” I think a Wikipedia that gets linkjuicde from millions of web “voters” but doesn’t give any linkjuice back is broken.

  18. Ben

    Matt,

    How do you feel about using nofollow on internal pages to channel pagerank? Does what I just said even make sense? My example is my site-wide link to my privacy policy… its easier to add a no-follow than to write the link in javascript if I’m looking to not pass PR. So, does no-follow make sense for internal linking?

    Thanks,
    -Ben

  19. Peter

    How on earth do you find these sites ??!!

    Anyhow, I had slapped no-follows on my forum as I didn’t want to risk linking out to bad neighbourhoods – I mean it’s hard to police a forum and it seems if one foul link slips through you’ve had it.

    During the big daddy my site went supplemental and I think this is all I did – don’t know really, but as soon I removed it (and sent an email to google) I was back in.

    So now I have all links open and try and police.

    Should I really use no-follow on new members or not do you think ?

  20. Mark Thomson

    I agree with Philipp on the subject of Wikipedia links. If the link has ‘aged’ then it is shown to be trustworthy and should be counted. People who produce wikipedia link-worthy sites should be supported and encouraged.

  21. Hi Matt,

    I’m seeing a number of people incorrectly using a nofollow value in the rel attribute for links to pages of their own site, where they should be using a robots.txt disallow statement or a “noindex,nofollow” value in a meta tag to accomplish what they intend to do – keep the search engine spiders from visiting those particular internal pages and getting caught in spider traps and not having those pages included in the search engine’s index.

    I’m sure that the idea of webmasters adopting the value more widely than they should be is something that the folks at Google have also considered – so that the “no follow” value is used for links on pages that are to be trusted, too. Doesn’t this have the potential to eat away at the efficacy of pagerank?

    While there’s a detailed write-up of the nofollow value on the official Google Blog, I couldn’t see any references to it in the Webmaster Info pages on Google. Maybe putting some clear information there on the use of the value might lessen some of the confusion I’m seeing over it.

  22. Chris Purcell

    I’ve posted my idea at http://www.usemod.com/cgi-bin/mb.pl?GoogleBitslap and suggested a solution to the obvious first objection (spammers attempting to subvert the system in revenge). Discussion would be most welcome.

  23. Matt,

    If I have an image linking to an internal page, and another link below that image linking to the same page, does it make any sense to slap a nofollow on one of those links?

  24. Matt, can you please tell me…

    Would you advise webmasters in general to put nofollow’s to internal pages, that really have no major significance ? So that the internal PR distribution is tunnelled much better to the pages that matter ?

    Would you consider that a spam ? Or an abuse ? Or a incorrect use of the nofollow tag ?

    Thanks.

  25. As for the posting links on wikis etc… I’m going to coin a phrase right now:

    Wikiteering – the process of adding ones links to wikipedia pages, or creating wikipedia pages for the sole purpose of gaining links.

    I like the way that sounds…

    Personally, I wouldn’t care if they nofollow or not. I have a link on a few wikipedia articles to my slang translator (oddly enough not put there by me), and they get me a LOT of actual human visitors. They’re far more useful than regular links.

    for spambots, I’ve actually had to shut down 2 of my sites because of them. One of them posted a comment on every thread on my site multiple times / day. It got to the point where he used up 10meg of MySQL database daily between my comment system and forum.

    My current blog hasn’t been affected yet (suprising why not, its a PR5) but it’s forum was on the list, and I gave up and took that down too. It was much easier to just say screw it than to bother with it anymore. It’s sad too, because I had 512 registered forum posters when I killed it. Owell.

    BTW, captchas aren’t the greatest either.. you’d be suprised at how many users can’t figure out what to do with it despite the “type this code here” instruction. I started logging improper tries on my company’s finance application so I could see how many times people typed it wrong, and if they ever ended up filling out the form or not. Most quit after 6 tries of entering their ATM pin and never end up submitting the form. I’d say about 5% of users.

  26. Nice Site. Can’t wait for the next update.

    Jost Joking!
    Have a great vacation

  27. Matt,

    What is the deal with sites using nofollows for in-site links? I’m seeing it all over the place, it’s really bothering me. Am I missing something? If it is only to stop crawlers from accessing the target page, wouldn’t a simple robots-meta-tag do the trick as well? Or are they just overdoing it?

    Thanks

  28. I think some of you are losing sight of what this is really about. It is not REALLY about optimizing your site or about how to make it possible for you to get more links. From Google’s perspective it is about improving search results by making sure that links continue to serve as reliable indicators of a websites value. When you have links on your site that you don’t approve or verify it messes with Google’s algo. The reality is that most site owners aren’t going to go out of their way to improve Google or work for the greater good of all internet users. There need to be disinsentives like the ‘bad neighborhood penalty’ or the fact that sites witout ‘no follow’ will get spammed.

    The wikipedia discussion is sad. Too bad this great resource cannot also help search engines find authoritiative websites. I bet google already disregards wiki links. Perhaps wikipedia could boost it’s integrity and value by adding ‘no follow’ but allowing it to expire after a 6 month approval period. A link sandbox, if you will.

  29. >>I’ve talked to a couple SEOs recently who said that they have a full-time person on their staff dedicated to scamming links from Wikipedia and wikis

    That’s Wikipedia’s problem, not Google’s.

    For the most part, I would think that Google would want to follow Wiki’s links. Wikipedia tends to be authoritative on more topics than anyplace on the Internet. In fact, if I were starting a search engine, the first place I would crawl for links is Wikipedia.

  30. What is your take on placing “nofollow” on links to PDFs?

    We religously place nofollow on all comment tags that link to a website. However, we, like the others above, wonder about placing them on “internal” links as well. Like a “disclaimer” page or “privacy policy”. Or is a NOFOLLOW NOINDEX meta tag better? We don’t want those showing up in the SERPs compared to our home page. :D

  31. > Would you advise webmasters in general to put
    > nofollow’s to internal pages, that really have no
    > major significance ?

    Word of warning: nofollow is also being adhered to by MSN and Yahoo and potentially others. When we analyze “nofollow” we should always ask for *its real meaning and what we can conclude due to it*, hoping that this is the common ground all search engines stand on when implementing nofollow algorithms, now and in the future.

    In that sense it’s a little unfortunate this attribute is a verb, because the verb implies a single action when actually several actions could be derived. For example, if the wording would be “unchecked” then it would be clear the link doesn’t count as a vote unless the search engine algo would decide “I don’t care if a link is unchecked or not, I have great other anti-spam tactics.” And why not leave it up to the search engines? I prefer to say H1 instead of saying “double-this-keyword’s-value-in-SE’s”, because H1 potentially can lead to different conclusions (e.g. render the font bold + big, or emphasize it on Text-to-Speech devices). I prefer to say “unchecked” because then tools could apply their own behavior (e.g. render the link differently, with a little warning icon — sure, they can do that now too, but it’s not strictly applying the “don’t follow” verb).

    It’s more powerful to let the SE decide on different verbs than to hardwire the verb into the attribute. The W3C knows this: separate functionality – layout – semantics/ structure/ content, and your tools get the chance to become more powerful in the future. Some members of the W3C agree with “nofollow” being a little unfortunate in these regards.

    All that being said, of course the “nofollow” was already existing, and it’s also important here to value existing norms.

  32. Limestone

    Forgive me if I’m being thick, but the oompa loompa site you referred to, if not set up as an example of scamming, surely is only there as a spam link repository.

    All the links that I could see were spam, and there appears to be no way for a member of the public to add an entry.

  33. Phil – We pretty much think the same on this and have been using nofollow on our own sites using our own logic, it might have not been a smart thing eh?

    AND what we all want is the same, rules that are clearly stated that we can all agree upon and follow (or nofollow). What is it going to be Matt?

    Spend some time typing phrases into google groups and you will see the same thing, webmasters and spammers all after the same fish. I do understand why Matt has to be tight lipped but boy this stuff getting frustrating for us all!

  34. Aaron i think the #1 rule of thumb would be: If you didn’t put the link there, and it didn’t go through some manual approval process, it needs to be nofollowed.

  35. Dave (Original)

    RE: “For the most part, I would think that Google would want to follow Wiki’s links”

    AFAIK they do follow links with the nofollow attribute.

  36. Mike

    Any chance you can point me to some documentation on this? I mean, ANYTHING?

    I still can’t believe that rpogrammers promote undocumented code as a “good” idea. Insane in the membrane!

  37. Hahaha

    The Oompa Loompa site looks like it has been rumbled. It was working OK a few minutes ago. Follow the link on the front page of the nuclearblender site and this is what you get.

    “Forbidden

    You don’t have permission to access /oompa/ on this server.”

    I agree with the other poster who said it was a disguised link farm.

  38. Dave (Original)

    Always wondered what happened to wee man :)

  39. Smart Matt :) says smart and do Smarter :P, Well matt I have very good experience and you can say bad experience with rel=”nofollow” attribute.
    You are smartly using it I am completly with you but it only works for google for yahoo and msn rel=”nofollow” does not work.
    Check this you will find your blog giving me the link are coming on the top :).
    http://search.yahoo.com/search?ei=utf-8&fr=slv1-&p=linkdomain%3awww.theseoguru.com

    Hope that you will not exclude me :( :)

  40. Matt,

    Very good to see that you´re adjusting your definition of links that require a nofollow atribute. From “any link that is paid for” to “any link that you´re not in full control of” is a correct adjustment.

    I always though that the idea that any paid link should have a nofollow atribute, was exaggerated. Great to see this adjustment just before you go on vacation.

  41. Great Article Matt, Loved it!
    hehe…

    Its pretty good idea, but as long as other search engines doesnt implement the same, spammers will still have a great time :) Anyways, a good initialtive!

    BTW, I dont understand why blogs should ask/display the home page! But for Wiki’s sure its going to help!

  42. Yaron

    In a similar matter to Aaron Pratt’s list, what about links from blogs (or magazines/news sites) which do change per post/page, but are to all those bookmarking sites (del.icio.us, furl, digg, etc…) ?
    These links are fully controlled and put there intentionally, but they’re also there because they’re technically useful in general, and not because they’re actually relevant to the content of an individual post. So they shouldn’t really get the same treatment as more relevant links from within the post.
    Is nofollow the correct way to treat those, so they won’t be treated and counted as real outgoing links by search engines?

  43. Harith (sorry I didn’t see your response earlier),

    I agree that people who are basically selling PR are abusing the system, but that’s not the case with everyone who accepts advertising on their site.

    What if I only accept ads from companies/sites I’d recommend to my audience? Why should I stick an attribute onto that link that tells the search engines that I don’t trust the target URL?

    Why should the average site owner, who probably understands on some level that links are good for rankings, have to know that advertisements don’t count as links? Sure, if they’re smart, they’ll advertise in places that are targeted to their audience, but that doesn’t mean the algorithmic benefit is something they shouldn’t want and expect. So whose responsibility is it to tell them the ads are only for traffic? Is it dishonest for a site owner to accept payment for an advert, put nofollow on the link, and not warn the advertiser?

    I think this is throwing the baby out with the bathwater. Some links deserve to “count” whether they’re paid or not. Some don’t. That’s for the search engine to decide, not me. If I want a link to count algorithmically because I trust it, I’m not going to tell the search engines not to count it. If they look at it and decide they don’t want to count it, that’s their prerogative and I have nothing to complain about. Similarly, if I stick nofollow on a link because I don’t want it to count, it’s up to them whether they’re going to agree that it shouldn’t count. Because it’s up to them, it’s just not my responsibility.

    Besides, Google’s guidelines tell us to create pages for visitors rather than search engines. So how do visitors benefit from putting nofollow on a link? If I do that, it’s for the search engines, not my visitors.

  44. Hi Matt – Joe Ward here from Crawlability Inc. I found your comments on rel=nofollow interesting… In fact, I’ve flog-ed you here:
    http://www.vbseo.com/f104/matt-cutts-talks-about-rel-nofollow-5017/

    This very post will serve as an example of linkspam interpretation. Do you consider my link (1) spammy or (2) a valid contribution by someone with related information on the topic? I’ve actually questioned that in my post. ;)

    We develop an SEO plugin for vBulletin powered forums. In our software, we have made extensive use of the rel=nofollow attribute. In many ways, I think that we set a great example of how to effectively use the rel=nofollow attribute in a fair and consistent manner – at least in the realm of forums.

  45. If it’s that valuable a link, and it’s that useful a resource, and it’s that useful to your target audience, why not just link to it for free?

    Matt’s bang on with this one. Paid links do not deserve any credit.

  46. Word of warning: nofollow is also being adhered to by MSN and Yahoo and potentially others. When we analyze “nofollow” we should always ask for *its real meaning and what we can conclude due to it*, hoping that this is the common ground all search engines stand on when implementing nofollow algorithms, now and in the future.

    Philipp, would the use of the nofollow attribute on internal pages, be considered a non-ethical or spam method ? Of course, if we look at the general ideea, even the word “nofollow”, as you say, is screwed up as far as meaning and deception goes.

    I myself answered to hundreds of different users (just starting to chew up some terms), where they thought that pages with the nofollow attrbute were not indexed at all. And from the word itself, that’s the meaning.

    My question to Matt : would YOU consider using nofollow on internal pages, an abuse ?

  47. Adam, I don’t sell links, so this is all hypothetical, but if I recommend a site to my audience, I’ll give them one link on one page. If the owner of that site wants to buy ad space, the link is going to be prominently displayed, labeled as an ad, and will appear on more than one page. It’s an advertisement. I’d charge for it.

    If I also choose to refrain from encoding a lack of trust into the link because it happens to be an ad from a company that I also recommend, why is that dishonest of me?

    It would seem to me that if I put nofollow on the ad, I ought to put it on that link I already had up because I want to recommend the site. After all, if I’m required not to trust them, then how can I trust them sometimes and not other times?

  48. I’m gonna have to go back to the roots of pagerank here and the google guidelines and say that while i like the idea of nofollow, I don’t think it’s the best solution.

    Here’s why:
    1.) it’s non standard code
    2.) it’s doing something non helpful to the user.. ie, it’s writing a site for a search engine
    3.) if a professor pays another professor to cite his research paper, it’s taken at full value. When mcdonalds pays michael jordan to endorse their food, people believe it.

    I know it’s tough, but we need to come up with a better algorithmic way to achieve the nofollow effect. It’s like asking people to turn themselves in when they break the law. Sure some will do it, but it won’t stop any of the bad guys out there.

    There’s patterns out there: comment spams tend to have forms on them, and mutliple links, dates, times, and even repeated links… same for forum spams.

    Paid links tend to be un-related, and use similiar text all around them, and come in blocks. Track the patterns, at the very least it will force the paid link sites to actually endorse the links to break the patterns.

    With that said, I still say I am against paid links (unless they bring in visitors).. however

  49. Adam, I don’t sell links, so this is all hypothetical, but if I recommend a site to my audience, I’ll give them one link on one page. If the owner of that site wants to buy ad space, the link is going to be prominently displayed, labeled as an ad, and will appear on more than one page. It’s an advertisement. I’d charge for it.

    That’s an easy enough situation to handle. Nofollow on all the links except for the organic one that would normally be there. If there’s a legit reason to link to them, then you could easily stick it in on a page or two of the site and nofollow the rest. That would distinguish the ad from the regular link.

  50. [quote]I myself answered to hundreds of different users (just starting to chew up some terms), where they thought that pages with the nofollow attrbute were not indexed at all. And from the word itself, that’s the meaning.[/quote]

    My gut feeling is that “nofollow” links are followed all the same, either directly (Google follows the link when it finds it on a page) or indirectly (some enabled the advanced, PageRank-checking features of the Google Toolbar which always submit a surfer’s URL to Google). The real action might be “noVote” and the real meaning might be “notChecked”.

    (Didn’t find the quote syntax linked, in case above turns out bad.)

  51. Yes, that’s an easy way of doing it so that Google is happy with my choice, but my point is that “nofollow” is supposed to mean “I don’t trust this link.” If I’m already linking to the site because I trust it, then I have no reason (apart from making Google happy) to say I don’t trust it on the ads. It would be a lie to say I don’t trust it, and I should think that doing so would call into question my claim that I trust it on the non-paid link.

    So I’m not looking for a solution here. I’m simply stating that if I want to recommend a site to my audience, I see no logical reason to put rel=”nofollow” on it just because it happens to be a paid link.

  52. Dave (Original)

    Guys! No-one HAS to use nofollow. The choice IS yours to make.

    However, it’s certainly a SMALL price to pay to increase chances of long lasting listings in the Worlds biggest SE.

  53. Dave (Original)

    RE: “So I’m not looking for a solution here. I’m simply stating that if I want to recommend a site to my audience, I see no logical reason to put rel=”nofollow” on it just because it happens to be a paid link.”

    In my mind it’s LESS logical NOT to use nofollow on a paid link. After all, you aren’t selling PR are you?

  54. No, I’m not selling PR. I’m not even accepting ads. But if I were, it would only be for sites that I would feel ok about recommending even if they weren’t paying me.

    [quote]Guys! No-one HAS to use nofollow. The choice IS yours to make.[/quote] Have you read or listened to Matt being interviewed by Mike Grehan?[quote]Buying links is extremely risky. It falls outside of our guidelines, unless you add a no-follow tag. And that’s a very simple way to say, “You know what? I only wanted the traffic. I’m not concerned with search engines.”[/quote]
    See http://www.clickz.com/experts/search/results/article.php/3605961

    I’d say that’s cause for a bit of concern.

  55. Dave (Original)

    RE: “And that’s a very simple way to say, “You know what? I only wanted the traffic. I’m not concerned with search engines”

    More reason to use nofollow then.

  56. Matt –

    nofollow is intended to mean “my site can’t vouch for this link”. It has some technical value as well: crawlers need not fetch a URL only to find a nofollow or noindex. And unlike robots.txt, nofollow can be applied based on content quality or other rules than just location.

    We have used nofollow tags on various parts of our site — pages we don’t think most users would find interesting if they were looking in Google, but which are good to have in the site as a resource (e.g. cables to hook up a TV, memory cards for a digital camera).

    We interpreted nofollow as a neutral directive; neither an endorsement nor a repudiation of the link. Have we inadvertently sent the wrong message?

    Do you have any comment on whether guiding a crawler within a site is an appropriate use of nofollow?

    Thanks –

    Tom

  57. Yes, that’s an easy way of doing it so that Google is happy with my choice, but my point is that “nofollow” is supposed to mean “I don’t trust this link.” If I’m already linking to the site because I trust it, then I have no reason (apart from making Google happy) to say I don’t trust it on the ads. It would be a lie to say I don’t trust it, and I should think that doing so would call into question my claim that I trust it on the non-paid link.

    That’s one interpretation of what nofollow could be used for, but certainly not the only one.

    What about areas containing potentially sensitive information?
    What about areas that involve a user login/password?
    Original content you may not want indexed in order to try and protect its integrity? (There aren’t a lot of people who do this, but they do exist.)

    Those are just three legit uses for nofollow that have nothing to do with trust or lack of trust. And you putting a nofollow attribute on an ad link (or any other link) doesn’t imply anything other than it’s not to be followed. You’re reading way too much into that one, dude.

    Like Dave said, it’s less logical not to use a nofollow attribute on a paid link than to use one.

  58. What about areas containing potentially sensitive information?
    What about areas that involve a user login/password?
    Original content you may not want indexed in order to try and protect its integrity? (There aren’t a lot of people who do this, but they do exist.)

    You don’t need nofollow for those. That’s what the robots exclusion protocol is for.

  59. Dave (Original)

    Not positive, but I think Google still follows links with the nofollow attribute. It’s just a sure-way to prevent linking to ‘bad neighborhoods’.

  60. You don’t need nofollow for those. That’s what the robots exclusion protocol is for.

    Those who have built anything requiring any level of security or confidentiality will be able to spot the flaw in that logic.

    You’re right: the robots exclusion protocol is there to allow webmasters individual page controls. But what if others choose to link to that content (#3 would be the example that I’ve seen most)? That link could potentially pose a problem if the webmaster took down the robots.txt file or for whatever reason a bot couldn’t find it. (Side note: that’s why I prefer to use the meta tag personally. Even if the robots.txt file can’t be found, it’s right there on the page for a bot to figure out.)

    Nofollow adds an extra level of security in those situations (albeit a weak one), and there’s nothing wrong with a little healthy paranoia from time to time. Never rely on one technological aspect, because when it fails, you’ve got no backup.

  61. “Adam Senour Said,
    May 16, 2006 @ 11:50 am

    If it’s that valuable a link, and it’s that useful a resource, and it’s that useful to your target audience, why not just link to it for free?

    Matt’s bang on with this one. Paid links do not deserve any credit. ”

    What? The internet is still a Hippy Place in your opinion??? Everything should be free??? What the …. ? If 100 sites ask for a link from my site,. and I believe 50 of them are great resources, but I don’t want to link to so many sites at once. Why not give the links to those sites that want to pay for it?

    We live in a capitalistic world. If something has value you are allowed to be paid for it. And if that gives search benefits to the person you sell it to, so what?

    Lets turn your logic arround :) If the sites you want to link to are willing to pay for it, why not ask some money for it?

    Show one offline magazine that does not contain any ads. If you can show one, check how much the magazine costs… :) Paid advertising on a website is not a bad thing on itself. Abusing it is a bad thing though. In the end it´s still the search engines that have to detect what are the bad links.

    Also consider that Matt said: “You are who you link to”. Doesn’t that still apply when you put the rel=”nofollow” attribute? To a visitor it makes no difference. What if a site for children puts a bunch of links to adult sites with the rel=”nofollow” atribute? Those links don’t affect the rankings?

    If they don’t, all search engines that use the rel=”nofollow” atribute have a big problem.

  62. If by turn my logic around you mean distort it completely, then you’re absolutely right. That’s exactly what you’ve done. You’ve distorted my logic to fit your own needs. (And I’m okay with that…happens all the time.)

    Let’s take your example of 100 backlink requests. If they’re all valuable resources of some benefit, then why wouldn’t you provide them to your users?

    You don’t have to list all 100 on one page. You can use recordset paging, categories/subcategories, or whatever technique you want to break those links up, but you should still provide them.

    If you decide to filter based on the bottom line, then you introduce a certain level of bias to those links since you may have elected not to link a valuable free resource over a less valuable paid one.

    And if you want to charge for that backlink, that’s your business. While I wouldn’t pay for one personally, I can understand why others do. Heck, I use AdSense on a couple of my sites, so I sell them. But the trade is a topical link (which gives legitimate, targetd traffic to businesses) and does not affect SEs or my site in any way. Go for it, as long as it doesn’t impact on the SEs.

    Now…why would that matter to me personally? It really doesn’t, all that much. The problem is how it affects the vast majority, which is something I think we all miss.

    Let’s take Wal-Mart vs. the mom and pop department store. Wal-Mart is a big tough corporate giant Goliath type with deep pockets. And the mom and pop is a David type with short arms that can’t reach their pockets.

    So…Wal-Mart goes off and buys ads on all sorts of websites, knowing that they’ll get link juice in addition to the direct traffic from the websites. Mom and pop can’t afford to do so, so they get buried by Wal-Mart in the process both in terms of advertising and in terms of SERPs, as Wal-Mart gains a buttload of link juice. And what happens? The guy who spent the most won, even though they may not have had the most relevant resource.

    And therein lies the major flaw in paid link juice providing search engine benefit…it potentially sells results to the highest bidder, which is what none of us really want. None of us are Wal-Mart types with the bling to go competing, so the field needs to be leveled.

    And, in terms of your logic of providing link juice, why would you? If they paid you for an ad, then you giving them link juice drains from your own. That makes your resource that much less valuable and harder to find. It also means that traffic counts to the OBL could become artificially inflated by less detectable bots that would crawl and visit the site, thus cheating the advertiser. So there’s an argument that can be made that a nofollow link can actually be more beneficial to an advertiser looking to gain targeted traffic.

    It’s okay to be capitalist and want to make money (we all gotta eat, right?) but for every action there’s an equal and opposite reaction. In this case, the opposite reaction doesn’t justify the action.

  63. By the way, Matt, speaking of nofollow: you know you’ve got some Dutch bot crawling your site and the links? I keep getting a bunch of referrals from it and it seems to be ignoring the nofollow directive.

    I’m not at home now, but if you want to email me at the email address posted here, I can send you some raw logs (when I get back home, I’m gonna block the stupid thing too…it’s a bandwidth whore.)

  64. Adam,

    I agree that an ad isn’t the same as a vote, even if the ad is for a site upon which the site owner has bestowed a vote. It’s a commercial link, and a search engine can’t be blamed for choosing to treat it as such.

    But no one has responded directly to my quote from Mike’s interview with Matt:

    Buying links is extremely risky. It falls outside of our guidelines, unless you add a no-follow tag.

    Would someone please clue me in as to what’s “risky” and outside the guidelines about this? Why is it my responsibility to tell the search engine an ad is an ad and I don’t want them to treat it as a vote? And what risk am I taking by failing to do so, even if I fail to do so because, whether I’m getting paid for the link or not, I have a lot of respect for the site advertising on mine, and I have no desire to pass on a message saying “please don’t count this link”?

    And that “little healthy paranoia” of adding nofollow to any links to a page that’s blocked by robots.txt and/or the robots meta tag? I’m sorry, but that’s just superfluous code bloat in my opinion. Google has always respected the robots exclusion protocol and I have never once had a problem with keeping pages from being indexed. Nofollow is not supported by the W3C, but robots exclusion is. I think it’s wrong to recommend using nofollow on top of the robots exclusion just to be on the safe side.

    BTW, congrats on the new gig. I’ve always liked your posts over at TW, and I’m sure you’ll do a great job for the big G. But I think you’re wrong here.

  65. Oops. Wrong Adam. So who do you work for?

  66. I think the problem with nofollow is that Google still follows the links, even if it ignores the link for matters of page rank.

  67. Dave (Original)

    RE: “Would someone please clue me in as to what’s “risky” and outside the guidelines about this?”

    If you link to a site that buys links it increases the risk of linking to a “bad neighbouhood”. While a seller may not be selling PR, the buyer often is trying to buy it. If they do that, they ARE outside the Google guidelines and LIKELY try other tricks to better their SERPs.

    RE: “Why is it my responsibility to tell the search engine an ad is an ad and I don’t want them to treat it as a vote? ”

    You can always NOT make it your “responsibility” and take the risk. The CHOICE is all yours.

  68. Oops. Wrong Adam. So who do you work for?

    Gun for hire. I have no particular employment allegiance as such, but I do have a small number of companies I deal with on a part-time or casual basis.

  69. “So…Wal-Mart goes off and buys ads on all sorts of websites, knowing that they’ll get link juice in addition to the direct traffic from the websites. Mom and pop can’t afford to do so, so they get buried by Wal-Mart in the process both in terms of advertising and in terms of SERPs, as Wal-Mart gains a buttload of link juice. And what happens? The guy who spent the most won, even though they may not have had the most relevant resource.”

    well,. Wallmart could do that, but why would they? They don’t need to do paid advertising for link juice (as you call it).

    and moms and pops still rank high, even though there are so many link builders out there. The thing with moms and pops is that they´re more specific and as far as I can tell, Google still gives the benefit to a niche site and not to a niche page on a walmart like site.

    Giving away link juice is not a bad thing. You chose who you link to and if you´re willing to link to a site you shouldn’t feel bad about giving away “link juice”. The last thing you want is a “dead end street” website.

    The rel=nofollow should not influence your decision on wether or not you would link to somebody.

    In my opinion, if you´re willing to advertise a site on your website, you should not be worried about your own search results. If you do, don’t accept that advertiser.

    Also, but very few seem to realize this, linking out helps your rankings more than it hurts, if you do it right of course.

  70. well,. Wallmart could do that, but why would they? They don’t need to do paid advertising for link juice (as you call it).

    The point is not whether Wal-Mart specifically would (interesting spelling, by the way), but whether or not someone with long arms and deep pockets would. Wal-Mart was used as an example company only, and nothing more. And as you said, they could if they wanted to. That’s a concern for any SE interested in relevant, organic results.

    Personally, I don’t think they would either…they’ve got no reason to. But what about a large company that has the resources and has suffered some form of a credibility or damage hit recently (Target comes to mind immediately, but I’m sure there are others)? What’s to stop them or any one of the other large companies to drop 7-8 digits on an aggressive linking campaign, buying links left, right and center, and knocking off the mom-and-pop? The BMW banning showed that large companies will spend money on SEM/SEO work, so it’s not that much of a stretch to suggest that they’ll try this next if and when they get the inclination.

    Giving away link juice is not a bad thing. You chose who you link to and if you´re willing to link to a site you shouldn’t feel bad about giving away “link juice”. The last thing you want is a “dead end street” website.

    How does nofollow do that, exactly? It’s not like the link isn’t there. A dead end only occurs when the user can’t go any further…and the user can still click the hyperlink.

    The rel=nofollow should not influence your decision on wether or not you would link to somebody.

    Why would it?

    In my opinion, if you´re willing to advertise a site on your website, you should not be worried about your own search results. If you do, don’t accept that advertiser.

    Why not? The higher you rank, the better value you’re providing to both yourself and the advertiser.

    Also, but very few seem to realize this, linking out helps your rankings more than it hurts, if you do it right of course.

    So you’re going to charge money for something you’re already getting benefit from both from an SE and user standpoint.

    There’s a difference between entrepreneurialism and greed, and you really would be well-served to study it in more detail and look at things from both sides of the fence. It’s amazing what putting the shoe on the other foot reveals sometimes.

  71. Rob Said

    The thing with Wikis is that the philosophy behind it leans towards the idea that users should be trusted unless they prove themselves dishonourable.

    Nofollow seems to me to be the opposite of that.

    Compromises do have to be made and if linkspamming is set to undermine the authority of a wiki then action should be taken. However I can see why wiki operators and users would be reluctant to implement nofollow

  72. I have seen people complain about Google’s link hunger for years,.. always the moms and pops and wal-marts like sites are used to proof that Google’s link hunger is wrong and that moms and pops suffer. But that´s all nonsense because the SERPs still show moms and pops.

    Matts definition on when to use nofollow has changed from “paid links” to “links you´re not in full controll of” and that is the best definition there is. If “you are who you link to” then you should be careful who you link to and if your site allows visitors to add links, the nofollow is a perfect remedy against spammers. But paid links are not spam just because they´re paid. They can be spam, but that has nothing to do with what they costed. In anyway,.. spammers are the last people that want to pay for links. They can get them for free so why would they even care about paying for it… :)

  73. Hi Matt,

    Good post. I am trying for months now to convince others at Wikipedia to re-enable the nofollow attribute. I started in February and spent in April a lot of time to write up something to explain what it is and what it not is and to argue against every single argument that was brought up against it.

    Here is the Main Thread at the Talk Page of the Wikipedia SPAM Project.

    http://en.wikipedia.org/wiki/Wikipedia_talk:WikiProject_Spam#How_to_save_hundreds_or_thousands_of_hours_by_spending_just_a_few

    If you have a Wikipedia Account you might want to stop by and leave a comment. Thanks.

  74. What about the crawling of these links? I’ve measured a while back how often will a URL that has been linked with rel=”nofollow” would be crawled. The results are very weird. It seems that Google is doing the most crawling on the nofollow-ed URLs. What happens to the content of these URLs? Why is it crawled anyway?

  75. “I’ve talked to a couple SEOs recently who said that they have a full-time person on their staff dedicated to scamming links from Wikipedia and wikis”

    comment: dude! thx for the idea! lol! no seriously who are those guys ;)

  76. I noticed that google displays a link to my site from here in one of my comments when I search for backlinks.

    So does this mean that google still follows and/or recognizes the link from this site to mine but does not credit it with a “vote” from this site?

    I was under the impression that it would be totally ignored, or at least not show up in a link: search in google.

    Just wondering specifically how that works.

    I hope you’re enjoying your vacation and Adam is The Man! btw :)

  77. Matt, a less spam-specific version of NOFOLLOW would be great. I have links on my site that perform actions (i.e. adding a product into the user’s shortlist). They really just refresh the page the user is currently on while performing an action. Those links are surprisingly successful – in terms of Google rankings anyway. They currently appear in the top 10 for most of our primary keywords in the Sites from Ireland on Google.ie! The problem is that I don’t want users arriving on the site to find that they have already added a product to their shortlist. I want to tell SEs not to follow these links, not because they aren’t trusted, but because they DO things. You can imagine how bemused our users are when they arrive at our site from Google to find they are half-way through a purchase process.
    I know there are alternatives to the link/action approach we have used (like postbacks) but all of them seem to screw up browser navigation.
    If I use NOFOLLOW, will this have any negative impact on any aspect of my indexing/ranking/PR?
    Thanks!

  78. Pavel

    Hello Matt,

    Knowing that this propably won’t be answered here anyway, I still want to ask you something. I run a website with a trading script and in my google sitemaps I see lots of errors with files like for example

    http://www.mydomain.com/cgi-bin/trading_script/out.cgi?id=20&trade=tradingpartner.com

    My server is up and running all the time, I guess it’s the trading partners site which makes the problems.

    So, I think google sees these links as internal links although they open external sites.

    Would it help to add a nofollow tag to these links, or does it harm a website when a link which google sees as internal has a nofollow?

  79. Although the ‘no follow’ attribute hasn’t been able to stop content spamming on public areas like blog comments, trackbacks, and referrer lists it has managed to reduce it to a certain extent. One of the reasons could be due to the use of automated programs by spammers which enable them to post thousands of comments instantly which are targeted at a vast multitude of blogs.

    Apart from the no follow attribute the webmasters also use spam deterent techniques such as creating captchas, requiring comment makers to register and pre-moderation of comments.

  80. Great Information Matt. Please let me know one thing, does Google count simple url of a website as back link.

    For Example,

    I placed a url http://www.abc.com/ on some relevant website, but what those people did is they did not hyperlinked the url. Will this count as a backlink for http://www.abc.com

    Thanks in Advance.

  81. Hello Matt,

    I’ve used the no follow tag in order to control the pagerank leaking out. Can you confirm that using the no follow tag on some links within a page would increase the pagerank leakage to the page links that are not flaged by the no follow tag?

    Thanks

  82. I personally prefer to decide myself if someone should get a real link or not. Why shouldn`t I vote for comment links, if they are good?

    I like the idea of handling links like that:

    - Comments that obviously have nothing to do with my content will be deleted as soon as possible. That`s why I would have deleted all of the above mentioned examples which look like spam for me. Comments like that are not rearly helpful, neither for me nor for search engines nor for my visitors.

    - Helpful comments and trackbacks with links to nice websites will get a real link automatically. I do not like the idea of giving a penalty to nice visitors only because of pagerank issues (if there are some). This should be good for me, good for search engines and good for the homepages of my visitors.

    - Good comments and trackbacks with links to bad pages – or trackbacks flagged with nofollow at the other side – typically get a nofollow link from me (too). These links are the links that rearly should be flagged as spam because they are spam.

    If you think it`s a good idea to work with links like I do, feel free to make use of the WordPress Plugin “Nofollow Case by Case“. This lets you selectively apply rel=”nofollow” on comments.

    Wouldn`t it be best for Google results if everybody votes diligent instead of giving penalties to all of his links, Matt?

    In my opinion nofollow should only be used for spam. Hasn´t it been introduced for that?

  83. Good post. I am trying for months now to convince others at Wikipedia to re-enable the nofollow attribute. I started in February and spent in April a lot of time to write up something to explain what it is and what it not is and to argue against every single argument that was brought up against it. Белая Церковь

    Here is the Main Thread at the Talk Page of the Wikipedia SPAM Project.

  84. Not positive, but I think Google still follows links with the nofollow attribute. It’s just a sure-way to prevent linking to ‘bad neighborhoods’.

  85. Hi all,

    Is there an answer to the internal ‘nofollow’ question? I was recently introduced to the concept and I’m curious, like others, to know if applying nofollow to internal links has any benefits, in terms of ‘channelling’ PR or just by speeding up the indexing process when the googlebot comes to visit.

    Also, great article in .net magazine, Matt. Thank you.

    Thanks.

  86. SEO

    yeah… those stupid wiki’s!
    wonder what happen if we all use nofollow to them?

    Have a good one.

  87. I predict some people will now, in return, stop linking to Wikipedia, or “nofollow” their links to Wikipedia (following the argument that if they don’t trust their own system, we shouldn’t either, and also following social etiquette – returning a disfavor, so to speak

  88. Hi,

    I’m linking to video posted by other people on youtube.
    Because those videos are played quite often, I’ve got listed
    in the top 5 of those video pages in youtube.
    Those links in youtbe have the tag nofollow so far as I know.

    But to my surprise i can fin them in my external links in Google Webmaster Sitemap. Does this mean that the tag nofollow in youtube is not recognized by google bots ???

  89. Thanks for the clarification on this issue. It seems to be something that isn’t going to go away and we need to make sure we address it. Thanks again. Scott

  90. Hi. Just because a backlink is displayed does not mean Google gives you a credit for that link. There are many such backlinks but would not do anything to your PR

    tks!

  91. Mark

    Hi Matt,

    As a directory owner who sells featured links would be be acceptable to Google to use a comment tag to show links are featured?

    For example:
    <– Begin Featured Links –>
    Featured Link 1
    Featured Link 2
    Featured Link 3
    Featured Link 4
    <– End Featured Links –>
    <– Begin Standard Links –>
    Standard Link 1
    Standard Link 2
    Standard Link 3
    Standard Link 4
    <– End Standard Links –>

    Or this totally unacceptable to Google and the nofollow has to be used?

    Thank you for your response in advance.

  92. Thanks for the clarification on this issue

    I was often confused what to do of No follow links..

    Thanks for clearing me out!

    Ash

  93. “Anytime you have a user that you’d trust, there’s no need to use no-follow links.”

    What about outgoing links on my resources page that I trust but don’t get a link back. If I put a no-follow on these, does this have a negative effect on these sites?

  94. Hello,

    What about adding the rel=nofollow tag to the target URL of forms.

    http://www.whistlerplatinum.com/

    If you look at the newletter form on this page above it leads to an external email list service. I do not want to bleed pagerank into this external website so I need to add the rel=nofollow tag to the form action URL. Is this possible?

    Matt
    http://www.candat.com/

  95. The no-follow tag is an interesting one. The Google webmaster’s area shows me that a number of sites are linking to me (blogs). These are definitely nofollow links as I double-checked the source-code.

    I wonder if google is simply saying “yes, we acknowledge here that site A is linking to you, but we are not crediting you for it”, of if they are saying “we are changing our policy towards no-follow”.

  96. Matt Cutts: Is This Article True???

    Every homepage has a given amount of “SEO Juice” that is can share among sub-pages *on the same domain*. So one should “no follow” unimportant landing pages like Privacy Policy, etc.

    Is this right???

    http://www.evisibility.com/blog/no-follow-tag/

  97. As Paul mentioned – Google webmaster tools are including some ‘no follow’ links in the back-links section (also Yahoo site explorer). Why bother indexing these as backlinks if they are not going to pass any link juice or benefit a website in regards to search rank or passing link weight…

  98. Very clean information for me. I completely agree with your sentence: “In an ideal world, nofollow would only be for untrusted links.”
    But after reading all the content of this page can we conclude that nofollow links count as backlinks?

  99. Ah… we have one massive flaw here. Paid ads are not the same as spam. However we are now listing them as one and the same.

    If the “nofollow” attribute is indeed a “bad vote” for a website, then by running a legit ad on another site with a “nofollow” added you are paying for a bad vote?

    Shouldn’t there be a separate tag for paid ads? A “nofollow” for links that you can’t account for and a “sponsored” for ads that are paid for.

    That way the search engines should spider a website that has paid for advertising. After all, in the real world advertising is to get noticed! By anyone. However I agree that the link shouldn’t pass link juice, just a notice to be checked out. The “nofollow” should not be followed or pass juice.

    But seems that while Google want’s us to do its work it can’t even offer us the right tools for the job.

  100. Thanks Matt, I got here through seomoz. I have a question. What do you think about the practice of nofollowing certain pages in your site like: contact us, privacy policy, etc.?

  101. Hi Matts,

    Just a question. I implemented nofollow for all my sponsolinks, i ask with my Webmaster tools to reevaluate my website marche.fr. How many time get Google for reevaluate a website ?

  102. I think now I understand what is the real use for Nofollow. Thanks for your info.

  103. The nofollow tag is something that is supposed not to pass on link juice to the external link, but I notice that even Google seems to consider these links. For example, I am a webmaster for a few websites, and using the Google inlinks checker, I came across many inlinks with nofollow tags. Now, is there a difference between these two?

  104. I think that it is unfortunate that the good should be punished along with the bad. I have always wondered what the definition of spamming is. I can see clearly that advertising your camcorder sales site on a dating site is (or gag dating site) is inappropriate. But on the other hand these indiscriminate spammers have created such a sensitivity that it is difficult to discuss a site that you really believe in, in related and appropriate forums. NoFollow is definitely a good option to have at hand but it does have the same issue of not being able to differentiate between good links and bad links.

  105. Marco Demaio

    You say: “The best-known use for nofollow is blog comment spam…”
    But why then not removing totally the link?!
    Why does the editor let a spam link to be in its blog.
    Can’t Google see it’s crap, why doesn’t it ignores totally pages packed with links?!

    GOOGLE nofollow PARADOX:

    1) We start from the wide recognized assumption that links on website W1 that show other websites W2, W3, …, WN decrease W1 PageRank.

    2) Therefor the webmaster (or seo) of W1 will add nofollow to all links that point out to his website.

    3) But also the webmasters of W2, W3, …, WN will consider covinient to do the same thing of step 2 to their websites.

    4) In conclusion each website on the internet will become completly isolated with all link pointing toward it being nofollowed and all link pointing out of it being nofollowed.
    In this situation a search engine based on link popularity (as Google PageRank) looses all its efficacy because each website is a website that does not increase/decrease anyone’s else websites.

    Plz don’t misunderstand me, I love Google, and I think Google is not only a business but the only star remained on teh web against the “evil Monopoly”. Once was Yahoo, but now it’s too weak to fight, if we let the “evil Monopoly” ruling also the web and the servers’ world e we’ll see one day a web full of useless Flash or even worst Silverlight ‘silly’ animations.
    We’ll se a web with no document.write Javascript funcion because XML can not handle it.
    It’s not only important that Google survives, it’s important that goes on being a strong player.

  106. well like i said before on another blog google should just open up to do follow and anyone who spams should just be penalized and thats the end of it.

  107. I think that comments should always be no follow but it won’t keep spammers from commenting.

  108. I just read an article stating you said at the “SMX Broadcast June 2nd 2009″ that Google is now ignoring the nofollow tag. That it will now index and count Page Rank to all pages regardless of whether or not there is a nofollow attribute. Is that true?
    -DK

  109. I want to sell links on my website, but I don’t want to fall foul of Google. The link brokerage system I used automatically places the links on the site so I can’t stick in a nofollow. If I surround the ad unit with will that work?

  110. Hey I read this post about captchas and I put one on my site: http://www.cjeschke.com. It is just a box that you click that says I’m not a spam bot. Will that work? Please email me with your answer.

  111. Will

    Why is it that “no follow” tagged links appear in Yahoo site explorer as valid inlinks, if they are not supposed to be?

  112. Im never sure when to use a no follow.

  113. Haha these are some really neat pictures.

    I’m not sure why anyone would want to live in a shack if they could stay here for free.

  114. I can see clearly that advertising your camcorder sales site on a dating site is (or gag dating site) is inappropriate. But on the other hand these indiscriminate spammers have created such a sensitivity that it is difficult to discuss a site that you really believe in, in related and appropriate forums.

  115. Why is it that “no follow” tagged links appear in Yahoo site explorer as valid inlinks, if they are not supposed to be?

  116. Why is it that “no follow” tagged links appear in Yahoo site explorer as valid inlinks, if they are not supposed to be?

    A hyperlink is a hyperlink. Some are nofollow, some are not (a.k.a. “dofollow”). All are valid links (i.e. “deliver someone to the destination URL”).

    Yahoo site explorer is not the engine that power’s Google, either, so asking the head of the Google Webspam Team about the ins and outs of a competitor’s public backlink-checking tool seems a bit weird. Sort of like asking the head coach of a winning football team why the losing teams drink so much Gatorade.

  117. Why is it that “no follow” tagged links appear in Yahoo site explorer as valid inlinks, if they are not supposed to be?

    A hyperlink is a hyperlink. Some are nofollow, some are not (a.k.a. “dofollow”). All are valid links (i.e. “deliver someone to the destination URL”).

    Yahoo site explorer is not the engine that power’s Google, either, so asking the head of the Google Webspam Team about the ins and outs of a competitor’s public backlink-checking tool seems a bit weird. Sort of like asking the head coach of a winning football team why the losing teams drink so much Gatorade.

  118. I have a better understanding of no follow links now i have read this but still not 100% sure about them.

    If website A has a no follow link to website B, would website B still benefit from this? The only reason i ask is because a few competitors have a lot of spammy no follow links placed on high page rank sites that are irrelevant but for some reason it appears to benefit them?

  119. Why is it that “no follow” tagged links appear in Yahoo site explorer as valid inlinks, if they are not supposed to be?

Leave a Comment

Your email address will not be published. Required fields are marked *

*

If you have a question about your site specifically or a general question about search, your best bet is to post in our Webmaster Help Forum linked from http://google.com/webmasters

If you comment, please use your personal name, not your business name. Business names can sound salesy or spammy, and I would like to try people leaving their actual name instead.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

css.php