Pure White Hat? If Google cannnot even determine what an unnatural link is then what is the point?

30 replies
  • SEO
  • |
We are constantly hearing "natural links" this "natural links" that. "Stop link building and let links come naturally". We seldom question whether search engines (or even SE employees) can even tell what is natural.

Now within a week of each other comes these two SEO news stories that indicate to me Google is nowhere near close to understanding their own "natural" criteria

Google Says We Will Improve After Mistaken Unnatural Link

and then this today

Google Issues False Unnatural Link Warning Again: DMOZ

Bear in mind these are only the high profile cases. Its probably happening a lot more than these two times.

So the next time I build what some one calls an unnatural link I'll just respond that their natural link might be just as unnatural.

What do you think?
#cannnot #determine #google #hat #link #point #pure #unnatural #white
  • Profile picture of the author yukon
    Banned
    Probably a disavow link phishing gone bad, lol.
    Signature
    Hi
    {{ DiscussionBoard.errors[8488740].message }}
  • Profile picture of the author nik0
    Banned
    Not only Google has that problem, most of the natural link checking softwares that people use to disavow their links also have that problem.

    Homepage links on strong niche relevant PR6 domains are flagged by those softwares as unnatural. Why? Cause it leads to the actual url of the post which is PR n/a and thus they flag it, bunch of idiots who created those softwares.
    {{ DiscussionBoard.errors[8488839].message }}
  • Profile picture of the author Kevin Maguire
    Probably just a new guy, making an honest assessment in the quality of DMoz.
    {{ DiscussionBoard.errors[8488869].message }}
  • Profile picture of the author JimWilliams
    That's pretty weird, counting that Google has the best engineers. Though, lately Google's algorithm is totally messed-up. By following all the rules, nowadays you won't get the first positions (1, 2, 3). Well, that's quite a problem!
    {{ DiscussionBoard.errors[8488893].message }}
  • Profile picture of the author guypeleides
    If Google did 10% of what Matt Cutts says they do, SEO would be a much easier game.
    {{ DiscussionBoard.errors[8489507].message }}
  • Profile picture of the author patadeperro
    I would like to have more information about these cases, because I am sure they are the exception, not the rule, yes Google does not have 100 % accuracy, and may not be able to ever have it, but I think they are becomed pretty good at identifying patterns, so I am sure these two cases are following tha same patters that "spamy" backlinks follow.

    I am not defending Google or saying they never make mistakes, I think these cases are more an exception than the rule
    {{ DiscussionBoard.errors[8489990].message }}
    • Profile picture of the author Mike Anthony
      Originally Posted by patadeperro View Post

      I would like to have more information about these cases, because I am sure they are the exception
      Hey Pat....You are SURE?? based on what evidence? you work at Google? Let us in on the 411

      so I am sure these two cases are following tha same patters that "spamy" backlinks follow.
      Same question. Either you have some incredibly strong and wide evidence you are about to present or you are sure about what ever you believe.

      I am not defending Google or saying they never make mistakes, I think these cases are more an exception than the rule
      If it walks like a duck, talks like a duck and looks like a duck its probably a duck. Seems like quite a defense to me.

      Anyway These are just the ones that made the news. Unless you have been under a rock the last two years you should know that thousands of legit businesses claimed to have been tanked without doing spam links. This gives them more credibility.

      Lets call it like it is. there are plenty of situations where its a judgment call to say a link is unnatural. its not like the minute I put a link on one of my sites to another of my sites sit leaves special HTML markup.
      Signature

      {{ DiscussionBoard.errors[8490039].message }}
      • Profile picture of the author patadeperro
        Originally Posted by Mike Anthony View Post

        Hey Pat....You are SURE?? based on what evidence? you work at Google? Let us in on the 411
        No Mike I have no Google affiliation unfortunately, but I think these isolated cases bases on the following numbers:

        Google has more than

        50,000,000,000 indexed as 2012, lets pretend they have not indexed a single other page ever since for practical porpouses.

        The normal distribution explain us how probable an event is to occur, meaning that under mild conditions (no google around) any random variable (pages) independently drawn from the same distribution is distributed approximately normally, 2 estandard variations from the mean it is expected



        Meaning: 95% of the results will be corrected most of the times without a very tight improvement process, with no improvements, no algo updates, nothing like that.

        That gives you:

        2,500,000,000 that will have "mistakes" in their rankings under any regular circumstance, I would expect Google to be way more picky in their standards and decrease the variation as much as possible, furthermore I am sure their standard deviations should be lower (just a 6 sigma process means 1 mistake in every million results).

        I am not sure they are 2,500,000,000 of mistakes, I dont have access to those numbers, but even if they were it would be "expected"

        My point here is that we should expect some mistakes in their results as a natural part of the process that is why I am saying we would need to have more information about these specific cases that are shown, because obviously these are in the "errror zone" and those are the ones that we need to study, because then we can realize what patterns are not recognized by Google yet, and duplicate them with the very clear concious that we may be caught, but we may enjoy the benefits in the mean time.

        By the way it seems that God Over You is into "patterns" as well, it is one of the first ones I have read here talking about that, I did not know his work before, but I am going to follow him, it seems he is making some interesting tests.

        Respectfully
        patadeperro
        {{ DiscussionBoard.errors[8493488].message }}
        • Profile picture of the author Mike Anthony
          Originally Posted by patadeperro View Post

          Nice graph but meaningless because you are using it wrong. The deviations are unknown as wel as the mean is. Again youwouuld have to work fo rGoogle to be SURE as you claim. You are calling them isolated where thousands of business webmasters claim there are not that isolated at all. Furthermore you can rarely use statistics, deviations etc on human judgment calls. its not like raw data.

          Give a definition to a natural link where the link itself indicates its unnaturalness without a human judgment call? You can't.
          Signature

          {{ DiscussionBoard.errors[8493554].message }}
        • Profile picture of the author godoveryou
          Originally Posted by patadeperro View Post

          By the way it seems that God Over You is into "patterns" as well, it is one of the first ones I have read here talking about that, I did not know his work before, but I am going to follow him, it seems he is making some interesting tests.

          Respectfully
          patadeperro
          Appreciated. I have a background in it and just cross-applied my pre-existing knowledge.

          That having been said, applying the standard deviation model to this? I understand what your train of thought is... but when talking about a system of calculations that has purpose build variances built into it, I'm not sure that standard deviation is still meaningful.

          Would you agree that this creates an unpredictable environment?
          Signature
          Don't Know Me? - Read my interview at Matthewwoodward.co.uk
          http://www.godoveryou.com/
          {{ DiscussionBoard.errors[8493687].message }}
          • Profile picture of the author yukon
            Banned
            Originally Posted by godoveryou View Post

            Appreciated. I have a background in it and just cross-applied my pre-existing knowledge.

            That having been said, applying the standard deviation model to this? I understand what your train of thought is... but when talking about a system of calculations that has purpose build variances built into it, I'm not sure that standard deviation is still meaningful.
            I realize just because a patent exist doesn't mean it will ever be used, still I find this part of the patent you linked to funny & I do think Google has this built into their algo.

            Alternatively, or additionally, noise may be injected into the document's rank determination. This noise might cause random, variable, and/or undesirable changes in the document's rank in an attempt to get the spammer to take corrective action.
            I'm sure by spammer they mean anyone that attempts to rank a page, which would be anyone that owns a web page & cares about Google traffic (average website).
            Signature
            Hi
            {{ DiscussionBoard.errors[8493731].message }}
            • Profile picture of the author godoveryou
              Originally Posted by yukon View Post

              I realize just because a patent exist doesn't mean it will ever be used...
              I'm pretty certain they do use this one. Of course, not being a Google insider it's not as if I can every say 'Yes they use it' and mean it.

              Wouldn't life be easy if they were 100% transparent
              Signature
              Don't Know Me? - Read my interview at Matthewwoodward.co.uk
              http://www.godoveryou.com/
              {{ DiscussionBoard.errors[8493762].message }}
          • Profile picture of the author patadeperro
            Originally Posted by godoveryou View Post


            That having been said, applying the standard deviation model to this? I understand what your train of thought is... but when talking about a system of calculations that has purpose build variances built into it, I'm not sure that standard deviation is still meaningful.
            I was not trying to apply the standard deviation model to Google, that would be very naive, the point I was trying to make is that any system have mistakes, and in a non intervantional way, the mistakes you should expect are around 5%, nature it self behaves this way, if you get 100 random people from all over the world stadistically speaking you will have a couple of red heads, my point was that Google is not perfect, yes there are some mistakes and yes those are the ones we should focus on, BUT this mistakes are the exception not the rule, not because Google has been unable to clearly identify 100% of the cases it does not mean that they are unable to recognize "patterns" that was the point I was trying to make to Mike.

            Anyway, I am fairly concious that Google uses neural networks and non linear models, as you said it is embedded in any "machine learning process"... and that is the interesting part of these "mistakes" that claim just to have "white hat" intentions,we should study them to see if they are "really a mistake in the machine learning process" (Eureka!!!) or like most of the times they have a spammy pattern, easily identifiable by a machine even that it may be hard for a human, these sites that are ina "gray area" are under human review.

            And no Mike, I can't really say if a backlink is natural or not, nor Google as we can see, but they can identify patterns, so as you say if it sounds like a duck, walks like a duck and smells like a duck....

            I have been thinking about proposing an area in this forum way more geeky, with stats, cause effect diagrams etc... let me know if there are more people interested in this.


            Respectfully
            patadeperro
            {{ DiscussionBoard.errors[8493890].message }}
        • Profile picture of the author Kevin Maguire
          Originally Posted by patadeperro View Post


          Google has more than
          Geek ...
          {{ DiscussionBoard.errors[8493768].message }}
    • Profile picture of the author GGpaul
      Originally Posted by patadeperro View Post

      I would like to have more information about these cases, because I am sure they are the exception, not the rule, yes Google does not have 100 % accuracy, and may not be able to ever have it, but I think they are becomed pretty good at identifying patterns, so I am sure these two cases are following tha same patters that "spamy" backlinks follow.

      I am not defending Google or saying they never make mistakes, I think these cases are more an exception than the rule
      That is the longest sentence ever. Can you cut it up? I don't quite understand what you're saying.
      Signature

      RIP Dad Oct 14 1954 - Mar 14 2015.

      {{ DiscussionBoard.errors[8494067].message }}
      • Profile picture of the author nik0
        Banned
        Originally Posted by GGpaul View Post

        That is the longest sentence ever. Can you cut it up? I don't quite understand what you're saying.
        He's trying to say that Dmoz looks pretty spammy in nature and I got to agree on that. I never saw such a boring link directory ever in my life (design wise).

        If this wasn't a manual action by a new Google'er then it proves that the algorithm doesn't really weight in the fact whether a site links out to crappy sites or to quality sites.

        Some often heard argument that that degrades the quality of a private blog network.
        {{ DiscussionBoard.errors[8494133].message }}
  • Profile picture of the author godoveryou
    It's bound to happen more often with other examples.

    Machines are predictable, humans are not. The more Google uses human editing, the more mistakes you can expect. It's inevitable.
    Signature
    Don't Know Me? - Read my interview at Matthewwoodward.co.uk
    http://www.godoveryou.com/
    {{ DiscussionBoard.errors[8491386].message }}
  • Profile picture of the author KylieSweet
    Originally Posted by Mike Anthony View Post

    We are constantly hearing "natural links" this "natural links" that. "Stop link building and let links come naturally". We seldom question whether search engines (or even SE employees) can even tell what is natural.

    Now within a week of each other comes these two SEO news stories that indicate to me Google is nowhere near close to understanding their own "natural" criteria

    Google Says We Will Improve After Mistaken Unnatural Link

    and then this today

    Google Issues False Unnatural Link Warning Again: DMOZ

    Bear in mind these are only the high profile cases. Its probably happening a lot more than these two times.

    So the next time I build what some one calls an unnatural link I'll just respond that their natural link might be just as unnatural.

    What do you think?
    It doesn't matter what Google thinks if it is natural or not as long as you know what you are doing it will be fine. Just like me, it took 2 to three months before my website has the number-one spot in Google it's exactly like a business it requires investment for you to have long-term effect in SE rankings.
    {{ DiscussionBoard.errors[8491427].message }}
  • Profile picture of the author Fraggler
    Check out Google's webmaster forum and people are being told ridiculous examples of unnatural links. Real people sharing their website with their friends on forums or taking out classified ads to promote their website - for traffic, not links - on websites that don't even know what a nofollow link is.

    It's ridiculous if you're going 'white hat' and through natural circumstance or harmless marketing your links ends up on a site that Google doesn't appreciate. Negative SEO techniques just rub salt in the wounds. Google cannot detect intent: it's just a computer program.

    If you want to make cash off Google organic traffic then just go full black-hat. It still works and works fast, just don't expect to stay at the top for long and roll with the punches. Otherwise just forget about Google organic altogether and concentrate on other traffic sources.
    {{ DiscussionBoard.errors[8491704].message }}
    • Profile picture of the author nik0
      Banned
      Originally Posted by Fraggler View Post

      Check out Google's webmaster forum and people are being told ridiculous examples of unnatural links. Real people sharing their website with their friends on forums or taking out classified ads to promote their website - for traffic, not links - on websites that don't even know what a nofollow link is.

      It's ridiculous if you're going 'white hat' and through natural circumstance or harmless marketing your links ends up on a site that Google doesn't appreciate. Negative SEO techniques just rub salt in the wounds. Google cannot detect intent: it's just a computer program.

      If you want to make cash off Google organic traffic then just go full black-hat. It still works and works fast, just don't expect to stay at the top for long and roll with the punches. Otherwise just forget about Google organic altogether and concentrate on other traffic sources.
      Actually same applies to forum profile links, the worse ones ever.

      I don't leave my link here in my contact details to have a SEO benefit from it, only so that people can find my website when they click on my user name so again a 100% natural link that is caused by totally different reasons then SEO'ing my site.

      Now I ain't saying people have to stat blasting profile links with Xrumer or the like but just giving an example that the worse link type can still be 100% natural.
      {{ DiscussionBoard.errors[8492160].message }}
      • Profile picture of the author Fraggler
        Originally Posted by PotPieGirl View Post

        If you EVER build a link to your web page in hopes it will boost your rankings, it is NOT "natural".
        The discussion is actually about people who have links built naturally and often out of their control and then get penalised for it.

        Chasing authors to remove your link from an article they wrote about your site without any influence from you is ridiculous and contradicts the reason links are a measure of authority in the first place.


        Originally Posted by nik0 View Post

        just giving an example that the worse link type can still be 100% natural.
        Exactly. And a lot of small businesses or casual webmasters use forum links purely to share their sites with peers, with no knowledge it may hurt their site on Google.
        {{ DiscussionBoard.errors[8492219].message }}
    • Profile picture of the author Mike Anthony
      Originally Posted by Fraggler View Post

      If you want to make cash off Google organic traffic then just go full black-hat. It still works and works fast, just don't expect to stay at the top for long and roll with the punches. Otherwise just forget about Google organic altogether and concentrate on other traffic sources.
      MY Goodness....What has Matt Cutts and Google done? They have driven my friend Fraggler to full Darth Vader dark side mode.
      Signature

      {{ DiscussionBoard.errors[8492669].message }}
    • Profile picture of the author yukon
      Banned
      Originally Posted by Fraggler View Post

      Check out Google's webmaster forum and people are being told ridiculous examples of unnatural links. Real people sharing their website with their friends on forums or taking out classified ads to promote their website - for traffic, not links - on websites that don't even know what a nofollow link is.

      It's ridiculous if you're going 'white hat' and through natural circumstance or harmless marketing your links ends up on a site that Google doesn't appreciate. Negative SEO techniques just rub salt in the wounds. Google cannot detect intent: it's just a computer program.

      If you want to make cash off Google organic traffic then just go full black-hat. It still works and works fast, just don't expect to stay at the top for long and roll with the punches. Otherwise just forget about Google organic altogether and concentrate on other traffic sources.
      That Google forum is one sided anyways, not sure why anyone would go there for advice. It's like asking Colonel Sanders for his secret recipe so you can get some of his business traffic.
      Signature
      Hi
      {{ DiscussionBoard.errors[8493372].message }}
  • Profile picture of the author seekdefo
    The SERPs are messed up more today than 2 years ago.
    Signature

    Brevity is the soul of wit

    {{ DiscussionBoard.errors[8491769].message }}
  • Profile picture of the author A S M
    Unless you spam sites with the same anchor text over and over again in a short period of time ... I dont see how they can possibly determine what is natural and what is not... even then its still hard to determine.
    {{ DiscussionBoard.errors[8492144].message }}
  • Profile picture of the author godoveryou
    I was not trying to apply the standard deviation model to Google, that would be very naive
    Okay, that's what I was hoping to hear
    Signature
    Don't Know Me? - Read my interview at Matthewwoodward.co.uk
    http://www.godoveryou.com/
    {{ DiscussionBoard.errors[8493990].message }}
  • [DELETED]
    {{ DiscussionBoard.errors[8494249].message }}
    • Profile picture of the author nik0
      Banned
      Originally Posted by strategic seo services View Post

      I think that Google looks at link building patterns (ie: if you over optimize your anchor text) to determine whether the links are "natural" or "unnatural". For example, 1000 backlinks with the exact same anchor text showing up within a short amount of time is a red flag for Google.
      For the first time in my life I seem to agree with you

      Not sure how natural this would look though:
      • 1 Article Spun & Submitted to 20+ Social Networking Sites
      • 300+ Backlinks from Social Bookmarking Sites
      • 200+ Contextual Wiki Links
      • 50+ EDU Blog Comment Backlinks
      • 300+ Forum Profile Links from High PR domains (2 Tier Structure forming a Backlink Pyramid)
      • 20+ Web 2.0 Profile Backlinks
      • RSS Feed Submitted to Top 15 RSS Feed Sites

      :rolleyes:
      {{ DiscussionBoard.errors[8494437].message }}
      • [DELETED]
        {{ DiscussionBoard.errors[8494514].message }}
        • Profile picture of the author nik0
          Banned
          Originally Posted by strategic seo services View Post

          It must look quite natural to Google, as it is one of my best selling packages that my repeat customers continue to purchase and leave excellent reviews for. Results are what matter, not false assumptions.
          Any looked at for how long that works? For the rest same like god said, if it works it doesn't mean that it's good.

          I mean who on earth is using those 295 crappy bookmark sites that are mostly based on the exact same Pligg platform.

          I've seen quite a few penalized sites with a very similar back link profile to what you're offering, no offense, just hard facts.
          {{ DiscussionBoard.errors[8495096].message }}
  • Profile picture of the author godoveryou
    Just because something is working doesn't mean it "looks natural." It's just operating below a threshold required to create a negative automated response.
    Signature
    Don't Know Me? - Read my interview at Matthewwoodward.co.uk
    http://www.godoveryou.com/
    {{ DiscussionBoard.errors[8494651].message }}

Trending Topics