Does Google Filter Results?

[Garrett]

I've read many times over that Google's using a filter to help screen out results for certain search terms. GoogleGuy, Google's PR representative at WebMasterWorld, all but denies the existence of filters. Because of the continued filtering questions and adamant denials I decided to enlist some help from professionals and find out what a filter is, why Google receives filtering accusations, and what the implications of a filter would be.

A cautionary word - this report is based entirely on analysis by individuals who don't work for Google. They are all intelligent and certain of their observations, but this remains speculation. That said, let's dig in.

What is a filter? The most lucid filter description came from Daniel Brandt, of Google-Watch.org (http://www.Google-Watch.org). In his words a filter is "an algorithm that is applied post-facto, on-the-fly, to results produced by prior ranking algorithms, such that certain results produced by the main prior algorithms can be deleted or suppressed (a crude filter) or rearranged (a sophisticated filter) before they are sent to the searcher."

Carl Rajkowski, a prolific poster and SE watcher from the SEOChat (http://www.seochat.com) forum (and employee of iExplore.com (http://www.iExplore.com)) said, "a filter by nature would follow a strict set of "rules" solely based on stop words and not gauge a page by it's content or outside importance/relevancy/theme, similar to a spam filter."

Does the filter exist? Here's where expert opinion splits.

Andy Beal (http://www.websourced.com) said, "I don't believe that Google is using a filter at this stage."

Dan Thies (http://www.seoresearchlabs.com) agrees. "The only public statements we have from Google indicate that there's no secret filter, right? So, let's consider the two possibilities: 1. They're lying. There is a secret keyword filter. 2. They're not lying. There is no secret keyword filter.

"My bet is on #2."

Daniel Brandt, however, says "I call this a 'filter' because it looks like a filter, walks like a filter, and quacks like a filter."

Vaughn Aubuchon (http://www.vaughns-1-pagers.com), who declined to comment for this report, created the "Florida Update Dictionary Filter Chart (http://www.vaughns-1-pagers.com/google-florida-chart.htm)," which gives a visual representation of what he thinks may be happening. Brandt referred me to this chart.

So why did the filtering accusations arise? Rajkowski put it plainly: "it seems to me that Google has implemented a quality check but for a lack of a better term it's been dubbed as a 'filter.' The main object of the 'filter' was not to filter out sites at all but to simply make sure a site was relevant. Google wanted to know if you were telling the truth that your site was really about what your backlinks and content said you were about."

Beal put it a little differently: "what often happens is that certain keywords are more aggressively promoted by SEO's than others. The impact of an algorithm therefore appears to be exaggerated too. But no filters!!!"

Thies explained it this way: "What some people see as a filter, I see as Google attempting to understand the type of information represented by a page, and the type of information that’s appropriate for a given query. It’s not going to be as unsophisticated as 'commercial' or 'non commercial.' Google’s changes aren’t driven by SEO; they’re about Google’s users."

Brandt put it like this: "once your search terms are found in the dictionary (this is an oversimplification, but will do for now), then the pages returned by the search are analyzed for their 'over-optimization' on those terms. The use of the terms in the title, in headlines, in links (domain, path and filename) and in anchor text attract extra attention."

Notice that there are four different explanations for a similar phenomenon - the seeming adjustment of Google's algorithm for certain search terms.

Filter or not, what should webmasters do? Brandt wrote in his article that, "a small number of webmasters have claimed improvements after tweaking their pages to avoid the filter. They rarely want to say exactly what they've done, but it can be assumed that certain two-word terms that seemed toxic were broken up, or synonyms used, or variations introduced by using stemmed forms of those words."

Brandt's suggestion doesn't prove the filter's existence necessarily, but if you're still not ranking well, even after the Brandy update, be sure to try and rearrange your keywords on your pages or adjust your use of words if possible.

Rajkowski, who optimizes iExplore.com, said "I SEO a website that is SEOd from the neck down. Every possible SEO trick and tip has been implemented into the design. I'm SEOd but I was never "filtered" during Florida or Austin because my backlinks justify my relevancy."

His best advice then is for webmasters to get relevant links and not worry about filters.

[Garrett]

Here is the complete text of my communication with Daniel Brandt of http://www.Google-Watch.org.

>What is a filter, in search engine terms?

An algorithm that is applied post-facto, on-the-fly, to results produced by prior ranking algorithms, such that certain results produced by the main prior algorithms can be deleted or suppressed (a crude filter) or rearranged (a sophisticated filter) before they are sent to the searcher. The main characteristic is that it's done on-the-fly. PageRank, by contrast, is (or was, at least until last April) precomputed and cannot be done on-the-fly, because the computational overhead for an accurate calculation of PageRank is very substantial. If you do something on the fly, it means you do it not once per page per update (which is how PageRank used to work), but once per search. Google cannot tolerate substantial overhead for every search, because they handle 250 million searches a day.

However, if you are using semantic analysis the way that Google appears to be doing since November, then a certain amount of that analysis cannot be done until you are presented with the user's search terms. Of course, you only have to apply it to a tiny subset of pages (between 10 and 100) that are presented by the main prior algorithm, so that helps a lot. I call this a "filter" because it looks like a filter, walks like a filter, and quacks like a filter.

>What evidence have you seen that proves or disproves that Google's using one or more filters?

The fact that it can be switched on and off by using any of several search techniques means that it's a filter. On www.google-watch.org/scraper.html you can see "pre filter" and "post filter" results. All the techniques involve using search operators that invoke advanced Google search options. Some techniques are better than others. The original technique was to exclude terms that wouldn't show up on any page in any event. Logically, this exclusion should have had no effect: keyword1 keyword2 -qwzwq -wzqwz

That worked beautifully from Nov. 21, when I discovered it, to about Dec. 11, when it stopped working. Google patched the exclusion technique in hopes of preventing searchers from investigating the filter.

Then the next technique involved including terms that should always be present. This technique isn't as "clean" as the first technique. One way, and the technique used by the Scroogle Scraper, is to do this: keyword1 keyword2 +www OR +com

This means if the page itself, or any URLs or links pointing to that page, include www OR com, then the page is produced. When parsing a URL, the period in the URL is considered a word separator. Therefore, over 95 percent of pages that are of interest to Scroogle users will be included. If a page is a dot-net or a country domain, AND this page uses a subdomain rather than www in the address, then Scroogle cannot not show accurate results for such pages.

Another way would be to use any of these: allintext:
keyword1 keyword2
keyword1 keyword2 +the
keyword1 keyword2 +and
keyword1 keyword2 +a

Note that the last three would work only for English-language pages. Scroogle doesn't use them because ever since Austin, more Western Europeans have been filtered, and began experimenting with Scroogle to see what might be happening. The allintext trick might discriminate against external anchor text, which would skew that technique. As you can see, none of the "inclusion" tricks are as "clean" as the original "exclusion" trick that Google patched, but you use what you've got.

>In one popular forum Google downplays their use of a filter, and all but deny that one exists. Why?

The word "filter" is a negative, while other words used to describe algorithms, such as "PageRank," sound positive. Google's public relations people are not about to brag about their new "filter," but they continue to brag about PageRank.

Also, there may be legal implications. Consider the SearchKing lawsuit. They complained that their original rankings were suppressed. Google esentially admitted this, but argued that PageRank was their "opinion" of a page, and was protected by the First Amendment. The SearchKing case was dismissed. You you can see how the nature of the case would have been different, and far weaker for SearchKing, if their complaint had been that their original rankings from Google weren't high enough.

The use of the word "filter" implies that some sort of entitlement was withdrawn. So besides being a public relations disaster, it raises questions about legal issues. If Google admitted that they were filtering out certain real estate sites because they no longer wanted them to show, what would prevent an association of realators from filing a class action suit? (By they way, it's clear from looking at the http://www.google-watch.org/hitlist.html chronology of filtering from Dec. 6 to Dec. 12, that Google made a special effort to partially reverse the effects of the filter for real estate related sites. I believe that this was done because the evidence of filtering for real estate sites was especially strong, and real estate people are well organized and well financed, and more likely to take legal action.)

They blocked the Scroogle scraper on Dec. 7, but I moved it to a different server within a few hours. It's been on that server ever since. By December 20 or so, I switched from the exclusion technique, which was no longer working as of Dec 11, to the inclusion technique. I think it's clear that Google would prefer that people not use the word "filter" to describe what's been happening to the SERPs since November.

>What would be the significance of this filter? What does the filter mean about the Google algorithm?

I pretty thoroughly hashed this out in http://www.google-watch.org/fiasco.html and don't have anything to add that I haven't already presented there. Primarily I think it's a tacit admission that PageRank isn't working well enough to keep the index relevant. This is either because PageRank, over the last three years or so, has evolved into a spammer's paradise, or because PageRank could no longer be calculated accurately due to other reasons (such as the docID 4-byte overflow theory presented in http://www.google-watch.org/broken.html)

[Garrett]

Here is the complete text of my communications with Carl Rajkowski, of http://www.iExplore.com.

It seems to me that Google has implemented a quality check but for a lack of a better term it's been dubbed as a "filter". The main object of the "filter" was not to filter out sites at all but to simply make sure a site was relevant. Google wanted to know if you were telling the truth that your site was really about what your backlinks and content said you were about. So Google started categorizing sites and pages.

A filter by nature would follow a strict set of "rules" solely based on stop words and not gauge a page by it's content or outside importance/relevancy/theme, similar to a spam filter. Google on the other hand IS calculating many more factors before it decides to exclude a page from the SERPs not just simple filtering.

In the past a webmaster would simply need to scatter some keywords for relevancy and gain backlinks to increase PR. Easy tasks that fooled Google into thinking a page was relevant when in all truth it wasn't. What Google is doing nowadays is checking the quality of backlinks, keywords, site size and themes to make sure a page is what it and other sites claim it is regardless if it's SEOd or not. Google has the ability, since Florida, to determine what category a site and page fall into so calculating backlink theme and keyword theme isn't that hard anymore.

Lets take an SEOd site for example that has many backlinks. These backlinks were gained from natural linking over the years. 90% of the links come from sites and directories that are similar in nature. Google looks at these backlinks, decifers a "theme" or category of the backlinks and if the site category matches than Google keeps it. Take the same site with the same amount of backlinks BUT now these links come from high PR sites that aren't related and many links are reciprocated. Google can tell that the unrelated high PR links were bought and reciprocated links were traded and the natural links that are left don't justify the site as being related because the bought and traded links out weigh the natural ones so it's dumped. This is why many users thought that SEO was the cause of being "filtered" out because buying and trading links went hand in hand with SEO in the past.

I SEO a website that is SEOd from the neck down. Every possible SEO trick and tip has been implemented into the design. I'm SEOd but I was never "filtered" during Florida nor Austin because my backlinks justify my relevancy. The website may have raised an eyebrow and made people scratch their head as to why I wasn't filtered but a Google "related:www.iexplore.com" and "link:www.iexplore.com" search will show why. My backlinks say I'm relevant, my content says I'm relevant, my internal links say I'm relevant. I could most likely remove 75% of my SEO and still rank well because I have enough outside factors saying I am who I say I am. *queue Eminem*

A great idea on paper but not when implemented without warning caused unexpected results. Sites were bumped because they were following the old rule set and did only what was needed to make Google think the site was relevant and not taking it a step further and making sure Google knew it was. Big giant word of mouth corporation sites, education sites, portals and directories all stayed because of years of themed natural linking. Spam sites started appearing because of blog bombing.

The quality check was just set a bit high at first because if you did good on one factor you were still dumped if you got a bad factor. That's been fixed since Austin. Now, for example, if you have a lot of backlinks it's ok that most will have the same anchor text unlike before you got dinged for your main keywords because of over use hence the "vary your keywords" rule was born. You'd appear for similar words just not the one you were targeting.

It's all getting fixed slowly and as you can see by www2.google.com sites are back and the results look great. And it took was 4 new datacenters, 3 months and a slacker algo. :)

Technically Google can say they aren't using a filter because in all honesty they aren't. Just a quality check to make sure one factor doesn't out weigh another and if you're legit you can do SEO until your face turns blue.

Hope that helps and makes sense.

Carl Rajkowski
http://www.iExplore.com

[Garrett]

Here's Dan Thies's take on Google filters. To learn more about what Dan does, visit http://www.seoresearchlabs.com

Google has some data sources, like click-through tracking, which are only visible to Google. They're still making tweaks to the algorithm to improve results, and we will see changes when that happens, but it's possible that we'll see a weekly "update" or even rolling updates, as click-through data becomes a part of the algorithm. It is now impossible to reverse engineer the Google algorithm, because a lot of the inputs are data that only Google has.

Everyone is picking up every change in search results, and naming it as an update. I would like to name all of the Google activity for the next few months “Update BAU” because I think this kind of frequent change is going to be “business as usual.” The Google algorithm probably includes click-through data, which is being fed back into the data centers with some regularity.

Anyway, you ask “what is a filter?” You’ll get all kinds of different answers. My answer today would be that a filter is “a convenient way of explaining something you don’t understand.” There are lots of theories, and lots of definitions, but they are probably all wrong. The only public statements we have from Google indicate that there's no secret filter, right? So, let's consider the two possibilities:

1. They're lying. There is a secret keyword filter.

2. They're not lying. There is no secret keyword filter.

My bet is on #2. What some people see as a filter, I see as Google attempting to understand the type of information represented by a page, and the type of information that’s appropriate for a given query. It’s not going to be as unsophisticated as “commercial” or “non commercial.” Google’s changes aren’t driven by SEO; they’re about Google’s users.

[BOBW]

Hi Garrett
I just want to reiterate my contention that google is filtering out some results in some listings. It seems that these are keyword phrases where certain paid listings or pay-per-clicker are vying for position. Several of these paid sites have no metas, no site optimization, and seem to rate on some searches. I can only now float yet another idea that when the $ is tweaked up, the filter kicks into gear.

Take the word “county,” when used in conjunction with certain names, my site is in the 500+ range, almost suggesting a negative effect for the word “county.” When I omit the word “county,” I am first again. Now on individual municipalities, too small and numerous for pay-pers to list, I am first.

The kicker is that the winners on the “county” word in some counties, are the same two or three sites, that no optimization. The general buying cycles of the public are emphasized and exaggerated in a seasonal business, and now is the time to tweak it. The paid advertisers with no optimization are doing so, but on a limited scale. The “county” searches are the best bang for the buck. Now I was first in the dead of the winter before the $banger$ - I guess I will have to fork out the dough-ray-me…

Bob

[kissels]

I own chicagolandwedding.com which, until mid-January, did VERY well on Google. All of the sudden, it did not appear at all for the "chicago wedding" keyphrase (where I was previously #1), then returned at #485. So, I started watching these forums and e-mailed Google thinking that I had been somehow penalized. I received the EXACT same initial reply Carbonize received (see Topic "I just had to share this") and, after two more back-and-forth e-mails, was flat out told by the Google rep that he could help me no further. My reply to that received absolutely NO response.

So then it was ALL about PR... Well, someone explain how when my main page has a 4/10 PR and I (at least previously) was optimized for Google, appearing all over the place in the top 10. After the January (was that still Florida?) fiasco, ALL of my competitors remained at least somewhere in the first 50 results while I essentially, from a user standpoint (c'mon, result #485... might as well be dead), had disappeared. My competitors had PRs either the SAME or worse (3/10) and a similar number of inbound links (although, Alexa shows 110 links to my site, while Google used to show only 19 and after Brandy, only 6). I was anticipating returning to the top of the heap after Brandy, but alas, am now at only #295 for "chicago wedding" while my main competition has replaced me at #1.

So, then we're back to keywords, keyphrases (2-words phrases being the most recommended). I have played and played with the "Chicago wedding" phrase on my site, both in my meta tags and throughout the main page's content, keeping the terms together, separating them, etc... and still come up short in the SERPs. Interestingly enough, while I seem to be "penalized" my main competitor that is now coming up at #1 has his "revisit-after" meta tag set for "1 days" which, I thought, was considered spamming the engines. AND HERE'S THE KICKER... I am the ONLY one of my competitors who spends money on Google AdWords.

So which is it... is Google "penalizing" AdWords advertisers (lots of controversy and Google denials about AdWords payers being intentionally affected, but is seems to be a common thread nonetheless) in the free/main SERPs because Google knows they will pay to advertise (MAYBE a little paranoid) OR is some kind of "filter" in place to keep AdWords advertisers paying for results? Something else that would make any sense at all? How can they explain away the fact that I'm the only one of my competitors paying for AdWords AND the only one so harshly affected by Florida and still NOT fixed with the Brandy update?

WebProNews' Newsletter today deals with speculation that Google uses "filters" to alter the SERPs just before they get to the user, thus allowing them to (if I understand it correctly) manipulate the results for whatever reasons suit their purpose. Good info... all should read. My point is this... While adamantly denying the use of "filters," couldn't Google essentially be using filters but have them so tweaked that they no longer fit the classic parameters of "filters," thus truth in the denials ... it's all samantics folks. Like Brandt said, "I call this a 'filter' because it looks like a filter, walks like a filter, and quacks like a filter." AND wouldn't it be to Google's benefit to use filters that keep AdWords and AdSense advertisers paying them for advertising... let's see... how would they accomplish this... oh, I know... kill them in the regular SERPs.

If this all seems ridiculous, please feel free to tell me so, but please... please... offer up something that makes more sense. At this point, I'll take anything!

[cbp]

So which is it... is Google "penalizing" AdWords advertisers (lots of controversy and Google denials about AdWords payers being intentionally affected, but is seems to be a common thread nonetheless) in the free/main SERPs because Google knows they will pay to advertise (MAYBE a little paranoid) OR is some kind of "filter" in place to keep AdWords advertisers paying for results?

Its neither. BUT, if you believe in conspiracy theories (Like Brandt does), you are welcome to believe it. Otherwise spend time actually understanding what is happening and work with that:
Read this thread:
http://www.webproworld.com/viewtopic.php?t=10802

Location sites (eg Chicago) have been more affected possible due to the semantic relationship (or lack of) to other words if Google have implemented a version of Latent Semantic Indexing - nothing to do with forcing people to buy adwords or fliters.

CBP

[I, Brian]

As I've said before elsewhere: if Google had implemented an expert system (such as Hilltop, or a variation of it) at Florida, then it would make sense to apply this expert set to a range of high traffic terms. Of course, high traffic terms are also often commercial, hence the origination of the accusation of there being a filter against commerical and/or SEO'ed sites.

However, the sad fact is that a lot of commercial sites - and even many supposed SEO sites - have extremely poor backlinking networks. As I keep telling people at IMR and elsewhere, it's fatalistic to rely on just a few handful of sites to provide the majority of your backlinks - it creates a narrow structure that is easily vulnerable to collapse at updates such as expert systems.

For proper backlinking, sites absolutely need to worry about the number of sites (on separate IPs) providing backlinks, rather than the number of backlinks they can get from an individual couple of sites.

[garv]

google definatly filter their results.

I was paying for Adwords, then when my site rose to the #1 listing for that specific term (it's adult so I can't post it here) that I had been paying for, I stopped using the adwords option. 1 month later my site was removed from google's database for that term. But somehow managed to keep #1 listings on Yahoo and Altavista.

After writing to them they sent me back a message saying that the site was still listed in google. The message is attached at the bottom.

If this is not shaping/filtering a search to create more money for their company I don't know what is.

regards
garv

PM if you want info about the site and search term


-------------------------
Hi Gavin,

Thank you for your note. Please be assured that your sites are currently
located within Google's search results and can be found at the following
links:

(removed due to adult content)

As you may know, we do not manually assign keywords to your site nor do we
manipulate the ranking of your site in our search results. Results in our
index change regularly based on ongoing, automated processes aimed at
improving the quality and content of our search results. Changes you
observe may include, but are not limited to, addition of new sites,
changes in the ranking of existing sites, sites falling out of the index
or getting dropped and sites' content fluctuating between old and new
content.

We realize these changes can be confusing. However, these processes are
completely automated and not indicative of wrong-doing or penalization of
individual sites. We currently include over three billion pages in our
index and it is certainly our intent to represent the content of the
Internet fairly and accurately. The ongoing changes you have observed are
part of this effort.

While we cannot guarantee that your page will consistently appear in our
index or appear with a particular rank, we do offer guidelines for
building a "crawler-friendly" site. You can find these guidelines at
http://www.google.com/webmasters/guidelines.html . Following these
recommendations may increase the likelihood that your site will show up
consistently in Google search results.

We hope the information we have provided above is helpful to you. Due to
the tremendous volume of information and help requests we receive, we are
not always able to provide personal attention to questions pertaining to
individual websites. If this response did not adequately resolve your
question, we hope that you will visit the webmaster section of our site at
http://www.google.com/webmasters/ . In an effort to better address your
needs, we've dedicated this entire section of our website to answering
common webmaster questions, listing Google's quality requirements and
recommendations, and much more.

Of course, we cannot anticipate and answer everyone's questions on our
website. Realizing this, we have also created a Google discussion group on
Google Groups where Google users and webmasters can connect to share their
vast knowledge and experience. You can access this group at
http://groups.google.com/groups?q=google.public.support.general . If you
have already checked the webmaster section of our website and haven't
found an answer, we encourage you to post your question to the Google
Support group.

Regards,
The Google Team

[kissels]

Okay... so maybe you missed the part where I explained that all of my competitors (who, incidently, don't advertise in AdWords) are doing just fine. In fact, my main competitor (who is spamming with a "return-visit" meta tag set to "1 days" and has a PR exactly the same as me 4/10 and similar backlink count - today Google shows his as 13 and mine as 9) has replaced me at #1 for "Chicago wedding" and I'm all the way back at roughly #295.

I would like to think that what you say holds true regarding geographically keyword-sensitive/dependent sites; however, this doesn't explain why all my competitors not advertising in AdWords are back at the top and I advertise extensively with AdWords and have really gotten slammed.

[cbp]

however, this doesn't explain why all my competitors not advertising in AdWords are back at the top and I advertise extensively with AdWords and have really gotten slammed.

How do you know its the Adwords? Have a look at were there backlinks are coming from (ie Hilltop algo theory; Themeing theory) have a look at the semantic relationships between words on the page (ie Applied semantics; Latent Semantic theory) - what are the differences between your site and theres?

I read a comment in another forum that upset a lot of people, but it said something like "you have got two choices - believe in the Google conspiracy or get with the program" - I think what they meant was do the above ... look at the theories that could explain the differecnces - look at your site and your competitors in the context of the more intelligent theories and ignore the conspiracies. make changes based on that.

Ignore the return in one day meta tag - its not spam and is 100% ignored by Google (ie its useless).
Ignore the number of links displayed by Google - they never show them all (they have only ever showed a sample).

CBP

[minstrel]

I have to agree with cbp here - I see a lot of people propounding the AdWords theory but the usual reasoning is rather like saying, "See? I went outside without a hat and caught the 'flu so that proves that getting chilled causes the 'flu".

You'll notice that the only one advocating for the filter theory is Daniel Brandt from GoogleWatch - the reservoir of Google conspiracy theories.

[DanThies]

Exactly what motivation does Googleguy have to lie? Make no mistake about it, if there is a filter as described by Daniel Brandt, Google's PR rep is lying.

It's true that many people have reported making (undisclosed) changes to their pages, and having them bounce back in the rankings.

It's also true that many people have reporting doing absolutely nothing, and they have also seen their pages bounce back in the rankings.

I work with sites where we do a lot of Adwords. Some of them have risen in the rankings, some of them have fallen. The same applies to sites where we don't advertise at all. Commercial or non-commercial. Directories and single page sales letters. Google changed the emphasis in their algorithm, to be sure, but they aren't cherry-picking specific words for special treatment.

Just because two or three things happen at the same time, that doesn't mean there is a causal relationship.

I ate Wheaties for breakfast, and my rankings fell. Did Google use the NSA's secret data feed from Kroger, cross-referenced with their Verisign domain name database, so that they can punish me for buying Wheaties? I suppose it's possible.

It's also possible that WebProNews and WebProWorld are giving a lot of credibility to a very biased source. I don't know how much of the "Google Watch Watch" (http://www.google-watch-watch.org) site is true with respect to Daniel Brandt's motivations, but I have to wonder.

[minstrel]

Just because two or three things happen at the same time, that doesn't mean there is a causal relationship.

I ate Wheaties for breakfast, and my rankings fell. Did Google use the NSA's secret data feed from Kroger, cross-referenced with their Verisign domain name database, so that they can punish me for buying Wheaties? I suppose it's possible.
Exactly - I just made a similar although less eloquent point in another thread (now merged with this one - see above).


It's also possible that WebProNews and WebProWorld are giving a lot of credibility to a very biased source. I don't know how much of the "Google Watch Watch" (http://www.google-watch-watch.org) site is true with respect to Daniel Brandt's motivations, but I have to wonder.
Well, there's no doubt in my mind that Google Watch has a vested interest in keeping the pot boiling when it comes to Google conspiracy theories - but it does make good copy and it usually does elicit some "excited" discussions...

[Mel]

Its sometimes hard to see the big picture when you are being squeezed, but consider for a moment why Adwords works.

It works because it puts your site in front of millions of targeted prospects at a time when they are interested in your product - these prospects are Google searchers who are there only because they believe that Google provides relevant results not because the want to see Adwords.

Now if the conspiracy theorists are right, Google is going to start delivering less relevant results in order to attract more paying Adwords customers whose results will then be seen by fewer and fewer surfers resulting in fewer and fewer adwords customers...

I'm sure at least one of those google PHds has figured this out.

[kissels]

Well minstrel & DanThies... I took a bit of what both of you said and ate a bowl of Rice Krispies out in the rain with my hat on.... My site is now, at least right at this minute back at #1 for "Chicago wedding" which, if you recall was the problem. Thanks. Actually, I "tweaked" the two keywords for about the forth time and finally found something the new Google likes.

Also... just so you know, I spent more hours than I can count doing comparison charts between my site and my competitors (PR, backlinks, keyword content counts & analysis, keyword title/description meta tag comparisons, etc...) and never did find anything substantive through my research to difinitively say why my site was the only Chicago wedding site that was hit so hard. With so much study and the only obvious glaring difference being my spending money on AdWords and my competitors not doing so, you really can't blame me for lending some weight to the AdWords theory. Just hoping I continue to rank at the top as before....

[garv]

My site went from #1 to not in the listings at all. I have waited since the start of September 2003 for my site to re-appear in the listings for the search term I use, e-mail after e-mail and they just kept telling me that my domain was in the database. It had a PR6 and over 50 linking pages, now dropped to PR4 and only 13 as I gave up trying after December. As stated is listed at the top of Yahoo and Altavista.

I never tried using Adwords again as I was to affraid of losing more business. I now just go for the search term and try to get to #1 this is the only site I ever used adwords for and the only site that has ever dissapeared. All the other Search Terms that I've tried for I've been listed within the top 5 and have stayed there.

regards
garv

[wolfy33]

We work for a company that owns the generic financial services urls. We have done well they were always in the top ten for their main keyword team i.e. credit cards, mortgages, personal loans.

Then Florida came and all two word teams like credit cards and personal loans went from the top 10 to not even in the top 500, even mortgages dropped to 7.

My point is we have the best urls the keywords are correct for the site and the content is all relevant to the product and site, we have some relevant links not all are relent yet we still dropped like a tone of bricks. On Google yet the bbc witch is powered by Google we were still number one for mortgages and 5 for credit cards I now they are a different algorithm.

When we first had Florida and you could put some rubbish after you search term we had 13 financial services sites in the top 10 when you did a normal search we only had 4 to me that’s a filter.

We are now working our way back up but things are still not the same.

[Mel]

Hi Chris, if you input terms appropriate to the finacial sector its no surprise if you get back financial sector sites even in those sites which showed with sites had dropped.
No doubt the Florida results were very poor and (IMO have now been recognized as such by Google) but the current google serps are very much like pre-florida serps, and in fact so are my rankings even though I have made no changes to my pages in an attempt to chase the florida update algo.

Try going back to the pages you were ranking in well before Florida. As an example I have three search terms that ranked like this before florida"

term 1 - #1
term 2 - #1
term 3 - #3

During Florida (up through the Austin update) my rankings were:

term 1 - not found in the firts 10,000 results
term 2 - not found in the firts 10,000 results
term 3 - not found in the firts 10,000 results

After the Brandy update (with no changes to the pages)

term 1 - #2
term 2 - #1
term 3 - #3

[AussieWebmaster]

Here is the complete text of my communications with Carl Rajkowski, of http://www.iExplore.com.






Technically Google can say they aren't using a filter because in all honesty they aren't. Just a quality check to make sure one factor doesn't out weigh another and if you're legit you can do SEO until your face turns blue.

Hope that helps and makes sense.


Okay so what you are saying is the smoke has cleared and optimization will work again though a quality check can still see you as over emphasizing a specific term but will help you for synonymns?