Forum Moderators: open
But my site still is listed, and i still get hits for all my other search terms, the much less important ones..
Have i been penalised for just that search term? I just dont understand it at all.
Many thanks,
I'm curious how you can differentiate this penalty (if that's what it is) from all the recent fluctuations?
After seeing homepages of mine drop and bounce right back, if I'd suspected something was amiss and made changes after I fell from grace, I'd probably attribute my comeback to those changes.
How long have you considered this a genuine penalty? Unless its pre-Dominic, I would have my doubts.
Which would seem to be the common case with a hypothetical site "Brandname Blue Widgets." It would make sense for the company to optimize its home page for their brandname, and the product they sold (assume here that blue widgets is the only item they sell.) Other sites linking to this site with "Brandname Blue Widgets" a lot would seem normal too. This filter would have lots of false hits unless it was well tweaked to penalize only cases that are so unusual they almost never would happen by accident.
I won't start the discussion again if such a penalty exists - I just want to fix the problem. I don't want to convince you - I have no problem if someone don't believe in this (simply forget this this topic).
(Have you read the behaviour which I had described in the other thread: A page is completly removed for a keyword2, while it's in the top ten for keyword1. Also, this page is in the top ten for keyword2 for a search in the directory. And so on ...)
... This filter would have lots of false hits ...
This is exactly the problem which I had. The offical name is "Companyname Widgets" and the domain is www.companyname.com. Most of the incoming links have the anchor text "Companyname Widgets" and the page was optimized for "Companyname Widgets". However, the page was nowhere if we were searching for Widgets until we changed the page.
Many innocent sites would be hit. Also big portal sites like in germany web.de would be affected.
Many innocent sites are already affected.
You don't know if this algorithm would consider web.de as over-optimized. If it's not considered as over-optimized there is probably no problem.
Also, there might be additional factors. For example, a spam report with that domain could be neccassary or pages with very high PR are not affected or ... (These are just examples. This is no theory.)
pages with very high PR are not affected
I have a PR8 site that IS affected - so far for another theory ... ;-)
Could you please be a bit more specific about the changes you made in order to "under-optimize" your sites? Would removing the H1 tag (if it repeats the anchor texts in use) be sufficient?
Peter
Are you sure the same page is being found? I don't mean the same url, I mean the same cached page.
I'm still seeing all sorts of weird things. Incorrect titles for pages, different versions of pages for allinurl to those being found in phrase searches, things changing daily, better positions in .com searches than .co.uk, old backlinks.
I know that a lot of pages/sites reappears in the SERPS after removing H1 tags.
However, this doesn't work in all cases. Some problems are caused by more than one factor.
MyWifeSays,
I know that Google has different caches for the same page (and sometimes I'm seeing some strange behaviour). But how should you know which version is used as long as the page doesn't appear?
My experience is that this isn't part of the problem.
The #1 listing is in England and is the manufacturer. They are staying at #1.
The #2 listing is fluctuating between my site and a .co.uk domain that hasn't had content for over 6 months and is now owned by the registrar (for the last 2½ months).
The #3 listing is fluctuating between a legitimate Canadian dealer and a page that was moved over 3 months ago.
So basically, when my site is not being ranked, the #2 and #3 results contain old, useless results. Whatever this "penalty", "filter" or "tweak" is doing, it's not producing beneficial results for the user in my case.
But I have a problem if you promote this theory because it is based on faulty logic.
"My site has disappeared. It has 'blue cheese' in the title tag. Therefore Google is penalising sites with 'blue cheese' in the title tag."
You have no proof of your argument. You say yourself that there are many variables and that other sites which disprove this argument (web.de) may not be affected because of as-yet-unknown other variables.
And sooner or later we have more Google mumbo-jumbo "accepted wisdom" which people view at face value.
Also, nowhere in the index is the normal www.domain.com only the www.domain.com/index.shtml in the off topic terms it does rank for.
It never showed the domain with the file extension before and ever since, the rankings are very strange and we are showing up well for terms that we shouldn;t be, showing up poorly for the one or two terms we should be, and always have, done well on.
Obviously traffic is down since the terms we're doing well on are mostly off topic.
Most observations I made then still seem to hold true - but factors such as competitiveness of the phrase (in number of results returned) seem to have been raised.
But, as others have pointed out, these are just educated observations - not proven fact.
That sounds far more plausible than a "penalty" applied to specific keywords and to specific sites, and certainly fits with Google's previous actions (in other words, to look for and apply the general rather than the specific).
That sounds far more plausible than a "penalty" applied to specific keywords and to specific sites, and certainly fits with Google's previous actions (in other words, to look for and apply the general rather than the specific).
It also might make sense with GoogleGuy's comments early in Dominic to diversify. Here is my my problem/question/conundrum.
My Company is called Blue widgets.com and naturally about 80% of the inbound links are with alt or anchor of >blue widgets.com<. The index page title is >Loch Blue widgets at Blue Widgets.com<. I have also lately been getting some links with anchor of >Loch Blue widgets<. Will this diversify my anchor text or reinforce the >blue widgets< part?
And would it help at all to diversify the internal backlinks? Maybe change the title of my index page to >Home<? Or perhaps >duilleag dachaidh<;)
But I have a problem if you promote this theory because it is based on faulty logic."My site has disappeared. It has 'blue cheese' in the title tag. Therefore Google is penalising sites with 'blue cheese' in the title tag."
stever,
I already had this discussion two month ago (in the original semi-penalty thread).
I never draw conclusions in the way you described.
I know that I haven't a perfect/complete explanation for the behaviour. I described major points which are part of the problem - nothing more, nothing less.
I know that this penalty exists, even if my explanations are incomplete.
My experience based on the examination of a number of different pages (i.e. on facts) and I know how the problem was fixed in all these cases. Of course, having studied about 12 different pages/sites and solved the problem is not enough for a final theory, but it's a good starting point. One can try to sharpen the picture by collecting more facts (that's what I want - instead of starting always from the beginning).
so the page doesn't even appear for certain searches, sorry i missed that point.
what have the positions been like after the page has been de-optimized?
I'm sorry if the analogy seemed a little bit basic - I was just trying to illustrate the logic.
In fact, since you mention the "semi-penalty" thread, I think that is a classic example of a whim becoming a theory and eventually taking on such a life of its own that people started to make new posts asking if they have the "semi-penalty".
At the best of times we are all blindfolded and trying to analyse the "Google algorithm elephant" by touch.
But now, with the elephant changing and sites appearing and disappearing within hours or days as datacentres alter their SERPs, it's an especially difficult time to attempt any kind of analysis that has validity. It is almost impossible to isolate variables and there is no stability to measure against.
re "semi-penalty" thread, I think that is a classic example of a whim becoming a theory and eventually taking on such a life of its own...."
I think you are right. But there is plenty of evidence that if you are optimising for a phrase, you can drop down the serps but not disappear, which suggests Google is identifying over optimisation and choosing to ignore the emphasis for the words that are over optimised. The same site can still do well for other phrases, hence the traffic still appears. This is a combination of seo techniques triggering the semi penalty. If the keyword is a brand name etc., any penalty will have little effect, but on a competitive phrase it can be dramatic. In this case, the sites that appear near the top probably have a more natural occurance of the phrase.
Figuring it all out is a wild goose chase, and the boring reality is (for us who, lets be honest, want to manipulate the rankings) is that we just have to knuckle down and produce the best sites we can.
the page appears for a search for 'keyword1' but doesn't appear for 'keyword2'. Of course, one can look at the cached page if you search for 'keyword1'. However, sometimes the cache shows different versions for different keywords. Therefore, you don't know if Google is using the same version if you're searching for 'keyword2'.
(However, this doesn't seem to be part of the problem.)
If the page reappears, it has the old position - apart from normal (small) changes caused by the updates. (At least this was the case in all cases which I studied.)
stever,
indeed, there were a lot of obscure theories in the semi-penalty thread. However, there are also a number of good posts based on facts (you can re-read the thread). Just because there were a number of very speculative theories doesn't mean that the whole discussion based on nothing. (You just have to filter out the facts.)
It's always hard if you seriously try to analyze parts of Google's algorithm. And you're right, it's almost impossible to isolate variables. However, the main goal of this discussion is neither a complety description of Google's algorithm nor a complete analyze of the keyword penalty - the goal is that pages/sites reappear. To attain this goal, (unfortunately) it's nessaccery (to try) to analyze the behaviour and collect experiences/facts. Of course, if there would be an easy way that a page reappears, people would simple do that.
I can understand you are asking for facts and proofs (especially if you are not affected and not involved in the whole discussion). Probably I would do the same thing under these circumstances. However, I'm beyond this point (I have already colleted as much as possible facts and carefully analyzed them). I just want to fix the problem.
soapystar,
interesting observation (I never heard about this so far). Unfortunately I can neither confirm nor deny this, because my experience mainly based on single keyword searches.
you can drop down the serps but not disappear
MHes,
there are a lot of people (I'm one of them) who can disprove this statement.
(I would be careful about drawing conclusions, if you are not affected by this problem.)
Fair point, I'll shut up.
(Mhes takes his ball home.)
