« | Home | »

The Art of SEO Bullshit

By theGypsy | May 19, 2010

Dumb asses and blow hards

Wow, what a work out going through my feed reader was this evening. Time and time again, from people I know, trust or laugh at… one and all left me spinning. You see, the SEO world seemed to be going through a period of blogging akin to watching the paint chips fall from the wall on a lonely long weekend. I mean it was getting fucking painful out there. Same old shit… mixed up and regurgitated to new piles of steaming uselessness. Over and over and over….

Well holleee shit. That’s sure changed of late. The new fad in town seems to want to baffle with bullshit instead of blinding with brilliance in the end analysis sadly. There is a distinct habit of SEOs trying to talk from the deep end of the pool while standing 8ft below the surface – bubbles… and lot’s of ‘em.

 

A geeks perspective

Now, let’s get things straight, I’ve been a huge proponent of SEO’s learning more about how search engines work… I started imploring such deeds to begin both 2009 & 2010. Hoorah!! Let us sing the praises by brothers and sisters! The SEO world is finally growing up.

But wait… something seems amiss. There’s a lingering odor of ignorance and… that other stuff… Oh yea, BULLSHIT. I am starting to wonder why some people out there are writing about these topics when they obviously don’t quite grasp the concepts of how things work. Could it be they are simply bullshitting to gain some elusive and all to often, meaningless, credibility in the search community?

Look Ma! I gotz skilz

There is no need to name names, because well, that’s bullshit too, but if you’re writing crap about how the SEO world is changing (aka MayDay bullshit) because of some patent filed back in 2004 you might want to keep reading. If you’re writing about the future of search inclusive of social or behavioural signals, it might be time to break out the Katana. If you’re writing about your amazing ‘testing‘ and pretty graphs… feel free to add your own ending.

For starters, don’t be fooled by the SEO Magic Bullet, (seriously, read that and come back). If you were REALLY a smart fucking search geek, you would have learned by now that;

  1. Technology from patents is generally implemented at the time of filing (ever heard of ‘patent pending’?). Furthermore, they are certainly evolved or abandoned in that 6 year period since being filed
  2. The FUTURE of search is more likely found watching the academic world that it is old patents

You see, one thing you will end up doing as a patent geek is researching the authors (engineers) that worked on a given patent. This will often lead us on a path from the school days where a pattern of thinking emerges. You begin to get a feel for the people that worked on it, their evolution and potential evolution of the technology in question.

This journey ultimately teaches one that many of the technologies we see are born in the academic world, not necessarily the search patent / research world. Dumb asses… stop glossing over with 6yr old papers and writing assertive ignorance for mass consumption.

 

Testing 1-2-3

Testing??? Holy shit. This one still amazes me… but is close kin to the one above. I stopped calling the mucking about I do ‘testing‘ a while ago. It really is nothing more than ‘research‘ if you ask me. After many years of reading and writing about search engines, one thing I do know, it’s abso-fucking-lutely impossible to isolate, in a real world situation with large enough data sets, any one signal nor set of signals to be able to really get much from any ‘testing’.

If we don’t know the majority of the signals and associate weights/thresholds/dampening, then how the hell can you possibly get meaningful data? Beyond that little inconvenience, we’d also need some fairly large data sets (different markets, query types, aged sites) to be able to really glean anything beyond anecdotal evidence. If you don’t grasp that… then stop writing about your shit for brained tests, ok?

I could go on, that’s likely a post on it’s own though.

 

Cut the crap

One thing I’ve learned being a student of Slawksi University, is the use of the terms, ‘may be’ ‘might’ ‘it is possible’ and the like. In short, there is nothing wrong with embracing search beyond top 10 lists. It is a glorious endeavour. What is problematic is there are those among us that either really don’t ‘get it’ or are pontificating for self glorification. Either way, shut the fuck up. M’kay?

I’d venture a guess that there are few even at Google has the keys to the entire ranking machine. More likely it’s a Manhattan Project approach with isolated teams working on various parts of the special sauce. While it can be edu-taining to try and guess at the various ranking signals, unlocking it is about as likely as ranking a site from comment spam links on Matt Cutts blog alone.

And if you’re NOT one of the bloggers out there blowing smoke out yer ass? Please be careful what you read – if it is stated as gospel, with no qualifiers, it’s probably… (all together now) BULLSHIT.

/end rant

No Bullshit? Then just Share it!
  • The Art of SEO Bullshit

Topics: Myths and Crap | 17 Comments »

Posted on 19 May 2010 by David Harry

Just a ranting and rambling search geek.... Web gypsy and renaissance man! When not losing what's left o' me mind here, I can be found blogging on the Fire Horse Trail and hanging wid me peeps at the SEO Training Dojo

has written 19 articles for SEO Bullshit

  • jimrudnick

    I too think that the 'recipe' for the google algo, is much like the Coke recipe….divided into three parts, each locked in it's own Atlanta vault, and each available only to a selected few. Time to make more coke? Each attends the 'mashing of ingredients' with their own offering….and NO ONE knows the whole ingredient list.

    Googles the same, I've always thought…..and nice to see others think so too!

    :-)

    Jim

  • LordManley

    Nice article, but I do feel that if you cannot get any benefit from testing then you aren't really trying hard enough.

  • http://www.seotunes.co.uk Matt Davies

    I don't think the problem is that testing is always inneffective, it's bloggers drawing concrete conclusions and passing them off as genuine advice based on limited testing. That same stuff is then regurgitated over and over until it's accepted as fact.

  • http://www.huomah.com theGypsy

    Hey Jim, good to see you as always. You are most certainly not alone there as I have yakked about that topic a few times with fellow geeks and if does make some sense that this is how things would be structured at search engines as well. It's not only risky, but unproductive in many ways I'd imagine.

    Careful your lordship. Once more, ANECDOTAL at best. Can you tell me all the potential signals used? Never mind their weights… Are you testing on large data sets of say 100 sites? Different ages, markets,query spaces? Then it really isn't worth doing more than going 'hmmmmmm'. That's the problem, most SEOs are so ignorant to how search engines actually work, they believe thier tests are meaningful. A dangerous prospect fer sure…

  • http://www.huomah.com theGypsy

    Hey Jim, good to see you as always. You are most certainly not alone there as I have yakked about that topic a few times with fellow geeks and if does make some sense that this is how things would be structured at search engines as well. It's not only risky, but unproductive in many ways I'd imagine.

    Careful your lordship. Once more, ANECDOTAL at best. Can you tell me all the potential signals used? Never mind their weights… Are you testing on large data sets of say 100 sites? Different ages, markets,query spaces? Then it really isn't worth doing more than going 'hmmmmmm'. That's the problem, most SEOs are so ignorant to how search engines actually work, they believe thier tests are meaningful. A dangerous prospect fer sure…

    Matt – precisely the problem that is happening. There really does need to be some qualifiers on these posts. 'While the data set was limited, here's some observations we made' – but that doesn't sound sexy and may not feed their egos as needed…hehe

  • LordManley

    I am testing across wide data sets of numbers comparable to what you suggest, but even then I agree that, for a lot of tests, the results can only be considered as indicative.

    We do a lot of eye tracking and user testing here at LBi, which is hugely useful, but we are obviously aware that, much like an exit poll, this only serves as an indication of what might be. This said, there are a lot of things which CAN be empirically tested.

    If I am looking to see whether a search engine gives additional weight to an H1 with a darker coloured text, even with hundreds of test beds I can only gain an idea. If, however, I want to see whether Google is following links from within noscript tags, I can absolutely glean a definitive positive answer.

    Of course, in that poor example, a false negative is available, but a false positive is not likely without actually making an error in testing.

    I accept that the test only shows what the search engine was doing at a given moment in time, within a given data centre, but none the less that can be valuable testing.

    We have a research team here who have a mandate which is purely to test and research SEO – they have the job titles 'Search Analysts', but we informally call them our Algo Team. The individual datum which comes out of that department is not often ground-breaking, but the accumulated knowledge acquired from continual testing and experimenting is very valuable.

  • http://www.huomah.com theGypsy

    Agreed… as you would likely know, after one does more and more testing, researches IR concepts… the more one realizes it is as U say, “a snapshot'. That becomes painfully obvious real fast. We've gone as far as having 50 people across the US (100+ internationally) run local search tasks. A ton of interesting data, but ultimately we're gleaning insights. We're building a few interesting program at the moment which is fun, and I encourage the curiosity.

    It is the way this BS get's peddled by many that seem to truly not get it. There are no qualifiers and it becomes gospel. It's simply irresponsible as far as I am concerned… meh… ohhmmm ohmmm… Inhale… Exhale..

    All better now… feel free to get in touch to compare notes any time U want though… Always yakkin algos in my world!

  • http://www.deepripples.com deepripples

    [applause]

  • LordManley

  • Pingback: Be careful what you read (about SEO) « This is Zemalf

  • http://www.raisemyrank.com/ Bob Gladstein

    I tend to refer to the sort of testing I do as “playing around” but I rather like “mucking about” too. I just try out a few scenarios and see if it leads to any potentially useful information — not how it all works, but whether a little something might be worth doing. And since I'm big on doing stuff for the user, the results of playing around usually don't matter a lot. If it's good for the user, I'll do it whether the search engines smile upon it or completely ignore it.

    On the rare occasion that I bring it up in public I always stress that everything I've done is Completely Unscientific, so if anyone bothers to look it over, I'm not really looking for a critique of my methods. But I get it anyway.

  • http://www.making8seo.com/ Adam J. Humphreys

    The Google algorithm changes upwards of 1-2 times a day. I doubt there's too many SEO's out there good enough to determine each and every change. What's really pissing me off lately is rejects claiming they're offering SEO when they've done absolutely nothing of the sort to promote a site. No, not even the basics, and making us all look like complete a-hats

  • http://x-pose.org/ Brant Tedeschi

    Not only is your post true, its entertaining to read the whole thing through. Nice work.

    I don't even read SEO articles because I equate the word “SEO” to bullshit. Therefore this website is the only one I need to read.

  • http://www.alliancetec.com Henry Gilbert

    what a wonderful blog. i like!

    but seriously can I guest post?

    There's a lot of SEO-bs practices followed by the herd – endorsed by the SEO-gurus – that I wanna rant about …

  • http://www.cheap-nikeshox.com/nike-shox-TL3.html nike shox shoes

    It looks good,I have learn a recruit!Recently,I found an excellent online store, thenrtrsmitters.com XX are completely various, good quality and cheap price,it’s worth buying!

  • http://www.nikeshox-outlet.com nike shox

    I don’t even read SEO articles because I equate the word “SEO” to bullshit. Therefore this website is the only one I need to read.

  • http://brand-cialis.net/ Brand Cialis

    A ton of interesting data, but ultimately we’re gleaning insights.