Google’s Glass Ceiling

A small “Google awakening” happened on Tuesday night. There seems to to be a “Google Glass Ceiling” in the serps for certain types of sites. Mileage might vary, but if your site is pulling more weight than Google thinks it should, there is mounting evidence that Google is throttling your organic SERPS traffic. It has been something that has nagged at me for sometime, but I was a little reluctant to risk the collective derision and laughter of the SEO community for bringing it up. But now I seem to be finding followers to the theory.

The awakening happened on a boat on the Thames. Maybe the Glass capsules of the London Eye provided the trigger? Maybe not. But there were 500 people on the boat – and pretty well every one of them is affected on a daily basis by Google… because A4U is by far the largest affiliate conference in the UK and – if Mathew Wood‘s peformance was anything to go by, will soon be the largest affiliate conference on the planet. These guys are good!

So I heard a person say “I can’t seem to break through a glass ceiling of XXX visitors a day from Google”. I then asked another well known affiliate, Loquax, who also admitted that even though his major site had dropped to 10th for his big one phrase keyword recently, his traffic had remained static from Google for nearly a year. Plot it weekly… flatline!

I talked to Jo Conner – with his inspired Can’tBArsed.com, a man who does far too much Tai Chi and far too little “real” work as I understand it. He said the same. I look at one of my own sites, Murder Mystery Games and I see the same. It seems to be that sites are “flatlining” on Google and although everyone seems to be settling at a different number, that number represents Google’s Glass Ceiling for the domain. You can push through it, but you’ll need to do something just a bit more… or at least a bit different.. to break the ceiling and take it to the next level.

 

Now I’m happy enough for Matt Cutts or Brian White or Adam Lasnik or John Mu to jump right in here and say “no, no, no Dixon – you must be dreaming” – but then it’s only going to take a few people to say “hey – yeh – that’s happening to me” for the idea to stick. At that point… if their really is a Glass Ceiling for some if not all sites, then one has to ask if Google are being fully transparent here? (Pun intended… an affiliate suggested I use it… I won’t name him).

 

It doesn’t – on the surface – seem to make sense. Google is all about pushing the best result to the users… aren’t they? But when your site is in that twilight zone of being well enough established in Google to stop looking at the daily rankings with fear and intrepidation, but not established enough to start getting “One Box” results (the ones where you get several links in poition one) on your almost generic brand name, then these are the kinds of sites that seem to be getting the Google throttle. Google knows there are several sites that may offer similar value to the user and it is in everyone’s interest (conveniently including Google’s) to give each of these sites a slice of the proverbial free action… in the spirit of “competition is good as long as it’s all done on Google”.

 

Not all styles of site face this ceiling. WordPress and the general blogosphere really wouldn’t notice this, as the blog content tends to work on link baiting and social network seeding rather than the traditional index to succeed. The result for bloggers is a far more volatile existence. There should be a saying – if there isn’t already – you are only as good as you last blog.

 

What worries me the most, is that if this hunch is correct, why has it taken so long to be “outed”? surely most of us look at traffic not rankings as a better key performance indicator don’t we? Maybe we jumped straight from looking at rankings before the Florida update many years ago all the way to looking at conversions from organic without stopping to look at the obvious middle statistic of the good old fashioned click?

25 thoughts on “Google’s Glass Ceiling

  1. Hey Dixon, I’ve kinda experienced the same stuff with a couple of my sites. What mostly would get me through these barriers is a Major DIGG, or an explosive amount of traffic for a few days, that could be something to work on I guess…

  2. >fully transparent pun
    Yeah, that would have been me, my missus can always see straight through me too!

    At the back of my conspiracy riddled mind I’d noticed my stats have stayed surprisingly consistent given all the things which can affect traffic on a daily basis so when Dixon asked a light definitely went on in my head – it needs a good de-frag in there.

    Great to see you again dude, great blog, we must do links sometime 😉

  3. Roy – I’m glad I’m not just dreaming it on my own. A Digg is something that might take the site into new territory… from old and stale to suddenly relevant content?

    Joe – I didn’t want to name and shame… 🙂

  4. Could it be connected to sites that are so indistinguishable that Google’s algo just plays with the serps because there is no standout winner?

    Over time you’d expect markets to be saturated with sites all ‘doing the right things’, all well linked, all just…..the same.

    Just a thought.

    d

  5. Surely would make sense as you suggest for Google to always show the ‘top sites’ in that particular search niche.

    However yes, I have for a long time always wondered about the ‘consistency of daily visit figures’. Over the longer term traffic steadily grows yes, I think with more and more people becoming searchers each day then any decent site should get this ongoing increase anyway but it is amazing to see daily visits within 10% of each other usually as oppose to much bigger swings.

    I also have noticed that these ‘ceilings’ (that works well as a vision) do seem to change as you break thru them. e.g you may average 50 visits a day for months, these daily figures would fluctuate only slightly.. 49, 48, 51, 50, 51, 48

    Then as you come across these jumps (for whatever reason) it may become average of 70 visits a day with figures like 68, 71, 70, 69

    This is a pattern I have seen ongoing.

    How can things be that consistant? I just dont see how it is possible without some mechanism in place as it would be if based on PPC budgets for example. That sort of ‘throttling’ would fit the picture pefectly.

  6. That’s what I was seeing too, Clive. However – someone has since suggested I get my maths head on and do some decent analysis. So I did. On the Murder Mystery Games site I compared all the Google traffic with the overall (Daily Unique) normalized traffic to look at an F-Test – which is the probability that the two data sets are not statistically different.. If Google WAS throttling traffic, then we would expect a HIGH f-test score (95% plus) whist if if it was really really close, we’d get a low score (5% or less). Here’s my data:

    _________________Google__All DUs__Normaiized Daily Uniques
    Week 28, 2008_____3,235___ 4,447___3178.298039
    Week 29, 2008_____3,104 ___4,346___3106.112723
    Week 30, 2008_____3,342___ 4,649___3322.668672
    Week 31, 2008_____3,059___ 4,279___3058.227414
    Week 32, 2008_____3,289___ 4,430___3166.148036
    Week 33, 2008_____3,095___ 4,319___3086.815658
    Week 34, 2008_____2,992___ 4,212___3010.342105
    Week 35, 2008_____3,022___ 4,367___3121.121551
    Week 36, 2008_____3,129___ 4,447___3178.298039
    Week 37, 2008_____3,121___ 4,515___3226.898054
    Week 38, 2008_____3,279___ 4,540___3244.765707
    Week 39, 2008_____3,464___ 4,801___3431.304

    Total_____________38,131__53,352___38,131

    F-test 0.518383903 or 51.8%

    Sadly, with a score of 51.8% my test is decidedly inconclusive. However the site doesn’t seem to be getting the same “ceiling effect” that I was seeing last year, so it would be interesting if a few people who felt they MIGHT be experiencing a glass ceiling were up to the challenge to do a similar formula (using Excel or similar) on your own data.

    Your dailies look very much like the sorts of similarities I had in mind when I was thinking about this. Maybe I should do that test on someone else’s data.

  7. Well it’s entirely possible given that Google have manipulated SERPS for years now. I suppose once the Google trend is out of the way and we all search via social groups, this won’t seem that important to us then. Great post though

  8. Hi

    Yes defiantly seen this, have seen two site stuck with the same traffic for over 12 months also seen one site were traffic jumps about every 3 months then has a drop off and then jumps up again.

  9. I’ve noticed several big sites that seem to have very similar amounts of traffic each month regardless of seasonal variations in that vertical. I’m not convinced that Google throttles the traffic but am open to the idea.

  10. Here’s a related point which may help confirm the basic idea of Google limiting the potential of a site (or a whole niche).
    As anyone with experience in p0rn can tell you, there’s definitely a glass pagerank ceiling for sites in the adult industry.
    It’s close to impossible to break the PR 7 mark and a PR 5 is rare. This is true for sites with hundreds or thousands of links.
    It’s like Google uses a curve when grading links for PR and adjusts it by industry type. This is all anecdotal of course, so just surf a bit and see for yourself 🙂

  11. I agree with this article. I started a site recently and I am pretty sure that google throttles traffic to just under 1000 visitors/day. Why? Because my position for the same keyword moves throughout the day. If anyone wants to see stats, I’m happy to show them.

    As to Google capping the PR of certain niches… I have proof of that. Adult is one industry, and there are others which I won’t mention here.

  12. Bit confused on the theory here (well it can only be a theory).
    Are you saying that after a certain amount of traffic google cap you.
    so in effect seo or search engine optimisation works as a negative by generating you traffic

  13. I am saying it might do, in certain circumstances. BUt as you say – it was only a theory. Alcohol was also involved and the post was written in October 2008… so I would find some more research (and some more recent research) before you add it to the SEO bible! 🙂

  14. I have noticed this myself, and the only reason I found this blog post was through a search “seo glass ceiling”. I have had various sites over the past few years that, for some reason, will not climb up past a certain point.

    Thankfully, these points have usually been #6/#7 so page one at least, but beyond that proves to be VERY difficult indeed, if not impossible.

  15. Sure that’s not only possible, but has been a factor in search algorithms ever since “Direct Hit” in the nineties. Google use click through rate as a major factor in the paid search rankings. Less obvious on the organic, to prevent the potential for abuse.

  16. This is all confusing to me. I have had a number of so called SEO experts tell me two different things. Everyone has different views and how Google works.

Leave a Reply

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