GoogleBowling - An Experiment in Link Sabotage

Story Text:

SEOChat members are testing the theory of Googlebowling: The practice of literally linking a competitors site out of the Googel index. They've linked 100,000 - 150,000 times from 40 or so domains to one site, with the same anchor text and await the results...

The big problem with this experiment is that the target domain is public. So, if Google are worried that people may feel that being able to damage websites you don't own just by linking could be damaging to them, they can just tweak it so that it doesn't happen...

The experiment would have been far more interesting had the target site not been named...

thanks to toddieg, the target domain owner for the tip...


This is going to be

This is going to be interesting to watch, thanks for the heads up.

Two Questions:

Will new domains have much linking power? Sounds to me like they're all new domains being used.

Anybody really think that Google will allow the site to be knocked out? You're right, it's too bad the domain being targeted was named.

Oh I believe it's possible

Oh I believe it's possible to get a site removed by power linking to it. I'm just not so sure in this case it will happen since Google's going to be aware of the experiment. And the domains currently pointing to the site are a write off now too for any other experiments (private or not). Tagged and flagged.

That's my opinion (I never seem to be humble lol).

>> since Google's going to be aware of the experiment.

...and that's another issue. If they let it work, it confirms that it IS possible to damage your competition. If they apply a handjob, they confirm that the manually twaek the index based on information gathered from forums etc.

Devil and the deep blue sea, eh?

Sheesh some people don't

Sheesh some people don't know how to conduct an experiment

Graywolf is right

No matter what happens in the experiment - it is totally flawed so nothing can be taken from it - it's a shame so much work for very little payback.

this isn't an experiment it

this isn't an experiment it is a publicity event, grab your popcorn and enjoy the show

I like "googlebowling" better...

it's to test the BLOOP theory...Backlink Over Optimization Penalty

That's a terrible name....


I believe it's worthwhile, even as a simple publicity/awareness technique. It's also a good way to see how GG responds to an issue like this.

>>see how GG responds to an

>>see how GG responds to an issue like this.

shall we take bets?


It's also a good way to see how GG responds to an issue like this.

That's the point, you won't know how they respond to it. If it fails, they may have intervened, if it succeeds, they may have intervened. Or, in both cases, they may have done nothing.

You have no way of telling.

We've been doing this kind of

experiment for some time, cept we didn't tell anyone.

It will be interesting to see what google does or does not do.

In my experience, the largest factor is the domain. For example, get bob over at searchqueen(you know who) to link to ya then link back and you can see how long it takes to tank a site, about 10 to 15 days... and it will only take a single reciprocal link.

We have added over 112,000 links overnight to one page, that page is currently #2 for its kw and as soon as gbot finds all the rest of the links it will go to #1, at least that is what we are thinking.

I actually tested something

I actually tested something similar but on a smaller scale, just after Allegra. I targeted some small venture capital companies for their own names to see whether that would recreate effects seen on other websites at the time.

It didn't - and to be honest, Google seems a lot more clued up on a whole range of issues than some people give them credit for.

When Google goes slightly screwy, it's very noticable - but I think Google do a very commendable job on trying to refine their results. I may not always agree with their ideas, but I've learned to develop a very healthy respect for how they think things out, and especially the general fairness in which they try to apply them objectively.

As for this experiment - if the targeted domain name is known, then the whole experiment is null and void.


...this is just the (public) control group *grin*


- lifting a #3 site to #1 as if it wasn't doing alright already... sheesh...

If GeoffreyF67 manages to get you guys to give him links for his autogen sites my bet is he'll be the winner here.

Sekret site

Just thought I'd mention that there are actually two sites that I'm attempting to bowl outta the SEs. One is public. The other is not.

I currently have around 300,000 pages waiting to get indexed so we'll see what happens when they do :)


Thanks geoff, that's good to

Thanks geoff, that's good to know.

Please post here when the results are in eh?

I certainly will. I expect

I certainly will. I expect it's gonna be another month or so before the results are in. Why I ever got in the SEO market where you gotta have patience I'll never know :)


The experiment is invalid.

The "west coast surfing" expression is very UNcompetitive. They need to try it in a highly competitive category.


Lots of people target uncompetitive keyphrases ;) To some, $1500 in adsense revenue a month is good enough... Others prefer to be #1..3 for an uncompetitive phrase over being #478 for [real estate]. I guess it takes all kinds to flood the web.

You seem to have missed the point

The purpose of the experiment is to determine whether a site can be moved down in the rankings through heavy linkage. They need to use a competitive search expression to test the hypothesis. Uncompetitive results are easy to manipulate.

Surely it either gets

Surely it either gets knocked out or not, why does competitivness matter?

Either it'll get nuked, or it will stay in roughly the same place right?

two different experiments surely?

is it possible to googlebowl a highly competitive result?

is it possible to googlebowl a less competitive result?

both would be equally relevant and could obviously take different numbers/proportions/lengths of time or one could be possible and one not, but they're both valid to try.....

If a site is already ranking

If a site is already ranking well for very competive keywords, then it's almost certainly established with a decent set of links behind it already. I'd say any attempt to influence the site is almost certainly going to face dilution.

Whether it impacts non-competitive areas remains contentious at best, but it may also be worth pointing out that just as sandboxing (the delayed impact of links) may work to prevent positive results, you may also find something similar for any sought-after negative results.

The hypothesis is that anyone can be Googlebowled

At least, the ORIGINAL hypothesis stipulated that anyone should be vulnerable to this kind of manipulation. Knocking someone to the bottom of the list for a non-competitive expression tells us nothing useful. I seriously doubt the site will be banned, so the best one can hope for is to get it dropped to the bottom of "west coast surfing" SERPs.

The real test of the idea would only come under competitive conditions. After all, in a non-competitive expression, it doesn't take much effort to shoot up to the top of the results. Many link-poor sites pass up link popular sites in Google's non-competitive expressions.

Hence, the only valid test is one where all other factors are equal: linkage alone has to be the determining factor. You only get that in the competitive results.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.