but has so many more relevant backlink than their competition that it didn't matter.
I am not so sure, although I said that their PR was high the 'followers' all have PR5/6. They seem to have 4000+ backlinks in G but how accurate can this number be? On Y! it says 150K+!
Following that time, many people took the words more literally and applied them to pages with high keyword density, long titles or just where the various aspects of the page were 'aligned' around the same keyphrase
that's what I assumed - OOP for me is both onpage and off page factors. Regarding the backlinks - 99% (link:http:www.domain.com/index.htm on Y!) - of the ones I could find were going to the homepage - the anchor however seem to vary slightly.
I'm intrigued by G displayed the meta description in the SERP for all the searches done. Was the search phrase actually on the top ranking page? If it was, could this give weight to the possibility that the phrase was ignored because of how it appeared on the page?
Yes the search phrase (3 words) appears twice on the page as I typed it in G.
Bad logic. I'm not say this is right or wrong, simply the conclusion is flawed.
Could you develop?
Gurantee you if you place every word on your page inside seperate H1 tags it won't do as well. Ditto if you repeat a word 500 times on a page, don't expect to rank for the word.
Well that wouldn't be considered as OOP in my books but blatant spamming. In the case of that company they use 13 H1's on their homepage (the one I looked at) most of them have 15 word/keyword rich sentences In this case again 95% of the words on the page (1500+) are links so I am not sure I agree with you.
Google's algo is a moving target and things change all the time but the importance of the link strategy seemed to remain a constant throughout most of the changes. Which led me to feel some doubt about the OOP.
I agree with you on both counts - links value and OOP existence
Over-optimization is a slippery term. Where does over-optimization end and mild spamming begin? I think no one would be surprised if one were to incur a penalty for mild-spamming.
However you want to call it - it seem to work for them. But I agree, as ciml said, it's hard to define OOP let alone tell when it becomes mild spamming.
Without seeing the URL its impossible to get into this disucssion.
Sorry but URL dropping is not permitted here. Moreover I warned you from the start that I wouldn't give away the link :)
isnt this site the huge site that was recently acquired by a huge print publishing company?
Don't think so - 500+ pages - UK based company.
And the one which is an authority site probabaly on a few thousand things, with thousands of employees, and massive optimization by all those employees?
Nope - it's only in one specific market but reaches almost all the segment of this market
but the age of the site, massive natural backlinks and large number of pages mean any penalty is balanced out by other factors.
in this case the site is 5 years old (domain first registered in 2000) first wayback page oct of that year.
There will be other over riding factors such as when the site is a "major player" with PR7, long established, etc. This is probably why the site in your example gets away with it.
I think so too - kewords on the page are all over the place.
on a sidenote: googled the domain owner and found one of her post(yes guys it's a lady - who said that the net was a man's man's world ;)) on a home working forum. She said that she just started the company and asked for advices - well I guess we're the one who need to ask her advices now.