| 12:27 pm on Jan 23, 2008 (gmt 0)|
Keep the big spenders down and they spend more on Adwords. I know I do.
| 12:41 pm on Jan 23, 2008 (gmt 0)|
Sometimes Google updates its filters causing them to be turned off for a few hours. During this time I've seen one of my pages go from #6 to #2. I think there is also a -10 penalty that acts in a similar way. If #12 goes to #2 (where it was before) for a short period of time then it sounds like a -10 position filter/penalty was temporarily turned off.
I've actually seen more of the -10 filter recently, especially after buying links with similar anchor texts.
| 2:51 pm on Jan 23, 2008 (gmt 0)|
> think there is also a -10 penalty
Maybe, considering that now several members think about -6 or -10 or
- 950 etc. etc., the "penalization" could consider percentage ( % ).
Maybe the penalization try to equilibrate the "weight" of acquired new links. In other words: depending on the numbers of acquired new links in a certain time, compared to your numbers of new links you "normally" get , if something differs from the "normal", google applies a percentage filter; the higher the gap, the bigger the "penalty".
| 4:19 pm on Jan 23, 2008 (gmt 0)|
A couple new observations:
1. Long tail impact. I'm very clearly seeing long tail terms stuck at #6 in addition to big money terms. Checking on the DC's without this filter shows these terms ranking #1 or 2, but they are #6 on DC's where the filter is in place. These are basically product type pages with obscure enough keywords that they have no adwords ads.
2. Position #6/11 ceilings. As search terms climb in the rankings I'm seeing them often hit ceilings at positions 11 and 6. Only after watching this for a while have I been able to determine this filter/penalty isn't just forcing terms down, but it's also preventing them from going up.
In spite of this, our traffic from google is still solid and we continue to rank #1 or 2 for a large number and variety of terms.
I continue to believe google is using this to limit the amount of traffic they send to our site, but I have no reason why they have chosen to do so. Maybe we are not spending enough on adwords. Maybe they don't like our linking strategies. Maybe they just don't want our business to become too dependent on free organic traffic. Who knows.
| 6:46 pm on Jan 23, 2008 (gmt 0)|
I clearly don't see any long tail terms stuck at #6.
| 12:45 am on Jan 24, 2008 (gmt 0)|
Here's another report of escaping the position #6 stranglehold - in this case top rankings returned after one of the main keywords was dialed way down from excessive repitition in image alt attributes.
A few other changes were made at around the same time: gained some new backlinks that didn't use the #6 anchor text, moved the brand name to the end of the title elements, and adding nofollow attributes for links to admin type pages. The site owner feels these other changes weren't the reason that the position #6 prison cell, and in this case I tend to agree, with one possible exception. One site escaped back in Decemeber and the only change there was getting new backlinks that used non-penalized anchor text.
| 12:55 am on Jan 24, 2008 (gmt 0)|
I like Hissingsid's four points:
|1. The term affected is the #1 big $ adwords term for your niche. |
2. The term has "Searches related to:" at the bottom of the SERPS.
3. If you use a plural word in the search term it pops back in to #1 or #2. For example search for blue widget #6 but search for blue widgets and back to #1.
4. If you reverse the term and search for widgets blue rather than blue widgets it pops back into #1.
In one case that is still under the position #6 spell, I see the reverse of his point number 3 - it's the plural that fell to #6 and #7, and the singular is still at #1 and #2. Reversing the word order (point #4) does escape the #6 box.
Now points 1 and 2 pretty much indicate that these are very common searches, and they may need some disambiguation - not that the Adwords price is directly involved. As I said before, I doubt that it is.
| 2:39 am on Jan 24, 2008 (gmt 0)|
|Here's another report of escaping the position #6 stranglehold... |
Ack. Hold on just one second!
As far as I can see, every site that was in #6 penalty is now OUT of #6 on many DCs (including google.com)
It's important to note that this happened whether
no changes were made or changes were made...
Let's not get back into throwing chicken bones about this.
Or drawing false conclusions about what's happening.
This is or was a test that as of NOW, showed no signs of being corrected by anything at all.
The only thing that changed, was Google fixing/re-adjusting whatever they were testing on those DCs.
| 2:48 am on Jan 24, 2008 (gmt 0)|
|I like Hissingsid's four points: |
Yes, I too can see the same.
| 4:51 am on Jan 24, 2008 (gmt 0)|
|showed no signs of being corrected by anything at all. |
I agree - we have not pinned down any factors that dependably get a url out of the position #6 stranglehold. There were some one-off reports of escape. Each of those anecdotes mixed in several types of changes at once, so there was no solid conclusion to be had.
All we know for sure is that Google did create this odd demotion and then hand it out to some top performing websites on some of their key searches. We can describe the effect, but we cannot find or fix the cause. And the effect may be thinning a bit right now - but it's not gone. The thinning seems to be Google's doing, not the webmasters'.
| 8:33 am on Jan 24, 2008 (gmt 0)|
My sense was you are absolutely correct Tedster. I will go one step further by saying that those who felt they had escaped the phenomenon for a large part were only seeing variations of DC's as information was flushed and added, which explains why the post was a see-saw about pos6 coming and going.
My sense was and still is that it was an error in testing that ultimately got looked at - for now. But it might point to a fundamental change in the way Google is ranking on those ladder steps - 11 and 6
| 11:00 am on Jan 24, 2008 (gmt 0)|
Hi, The 4 points I posted earlier have a number of explanations. I’m still thinking about this but here are some thoughts that might stimulate others.
1. The Google algo now includes a different type of semantics filter, possibly phrase based rather than word based.
2. The algo is possibly pointed at a dictionary of "selected" phrases.
3. They have possibly used the same or similar technique in selecting the phrases targetted as they do for the "Searches related to:" feature or maybe it is based on some form of broad match or some weighting based on ambiguity.
In my case the "Searches related to:" list is not what I would expect to see if it was based on an evaluation of the organic results. It includes "brandname blue widget" terms where the brandname does not operate in our “blue” market. Half of the related terms are odd ball and the other half are not the main related terms in our niche. They just don’t look right.
It does look right if you split up the term.
The brandname companies are all big market share companies (and big Adwords spenders) in the “widget” market but do not operate at all in the “blue widget” market.
The real word which I’ve replaced with “blue” is an ambiguous word.
Widget is not ambiguous at all, it can only mean one thing.
So if you focus on the unambiguous word and pair it with the big spenders on adwords and disregard the ambiguous word you get half of the terms in the "Searches related to:" list.
The ambiguous word in question is the same word we had a problem with in the post Florida period.
I’m not really sure what the answer is but I’m pretty sure that it has something to do with the ambiguity of one of the words in our two word term and the fact that the Google algo is now looking at that ambiguity in some way.
| 9:35 pm on Jan 24, 2008 (gmt 0)|
And now the ceiling re-rank is rolling back out through the datacenters...wow, just wow.
< continued here: [webmasterworld.com...] >
[edited by: tedster at 8:48 pm (utc) on Jan. 29, 2008]
| This 193 message thread spans 7 pages: < < 193 ( 1 2 3 4 5 6  ) |