Forum Moderators: open
"This thread is ONLY for serious, generic discussion of changes that you are observing with the new algo in this update. As in things like "Looks to me like PR is less important this month, and anchor text of inbound links counts more.", etc. How your site is doing has no relevance here unless you can explain why you think so in terms of a general algo analysis."
This thread is NOT for y'all to say how much you think Google sucks, or alternatively how great the new SERPs are. The idea is to pick apart how Google is working, and not to criticize their quality.
Has anyone here tried looking at their pages with the Adsense Preview Tool. You see by example what the Adsense algo thinks the theme of your page is. Point it to a non existatnt page within your site and you see what the overall site theme is.
I've noticed that in some instances sites that have an index.html page for which the Adsense tool cannot decern a clear theme and therfore sends charity Ads seem to do well if the overal site theme and page themes are right on topic. I've not looked at a big enough sample to conclude that this is a consistant trend though.
Ant comments.
Best wishes
Sid
If people ever wonder why Google Guy seldom gets bluntly specific, here is a good illustration.
Not only did the 64 results not move over to the other datcenters in the few days that he mentioned, they NEVER moved over. They are just gone.
Instead we are left with an algorithm that is being spammed basically every day with new insta-sites of near zero content, and no solid algorithmic assets besides (helloooooo 2003) piles and piles and piles of anchor text from related family sites.
The abandonment of the 64 data is bizarre since there seems no logical reason for it -- other than:
if webmasterworld members like something it is taken out and shot.
The current algorithmic mix is once again in love with anchor text and new/fresh pages. White bar sites are flying up the rankings in competitive areas. Often a top ten will feature three or four featherweight "sites" that never ranked before but are parts of large families of sites that throw up these insta-domains and link to them every time a keyword gets "hot" for one reason or another. At the same time, authoritative, content rich domains with plenty of anchor text from all the other authority domains that have been at the top of the serps are now mixing in with these psuedo-sies. Clearly linking from thematic sites matters far less now than a month ago. Authority means dramatically less (including generic, off-topic authority like pages on cnn).
Whereas the post-florida advice was to build deep content domains, this is most definitely not what is being rewarded now. Families of low content, keyword focused domains with thousands of interlinked anchor text pages are spreading like locusts.
It obviously depends on the keyphrases and topics people are watching.
64.* moved for me too.
Steve, are these strange results you are seeing in your own market or across the board?
I have to agree with Crush, my secondary sites have been dropped for no logical reason and once again they seem to be among a small select group that have been dropped. They all contain the "brand name" which is my main sites domain name in the text somewhere and link into my main site. There are other sites which have this brand name on them and links to my main site which are not dropped. Its all very difficult to understand.
Best wishes
Sid
>>but what is 64*?
They're refering to the datacenters that had an updated algo applied before the results were seen (or not...) on the datacenters www.google.com pointed at. One of the IP addresses is 64.233.161.99
For folks who didn't see 64* move over can you please say what broad category.
Like, financial, gambling, porno, b to b, consumer products, information, academic, travel, real estate etc. Its really difficult to comment if we don't know what area you are talking about.
If 64* didn't come over what did. What was different about what did come over and what was on 64*?
Sid
The abandonment of the 64 data is bizarre since there seems no logical reason for it -- other than:
if webmasterworld members like something it is taken out and shot.
I became intensely uncomfortable early on with Brandy, in that so many in here were singing its praises. It's almost like that made G too nervous. [I don't really believe they changed it for that reason however :-) ]
The sad thing is, while many in here liked it, I got that genuine impression that while some/many were just happy to see their sites back, on a deeper level, there was widespread acknowledgement that .64 was perhaps the best we'd ever seen from G.
Then, after a few days, something went askew again. It happened in so many categories that I could not possibly identify one. Most of our sites are more information oriented so we get a good look at a broad cross section. Nearly across the board, by about day 4 of Brandy, many of the very good sites in the top ten dropped between 2-6 places, sometimes more, and were replaced again by technically-relevant-but-not-useful drivel that remains in place now. It's so frustrating that they had it, then let it go, or changed it intentionally. It's like one of those movies where two long lost lovers walk past each other in a crowd but don't see each other.
Seeing the way they changed Brandy after the first few days made me really wonder how G judges quality. When a maker /provider of goods or services has a product in developmeht that is just right, and then changes it, it's a sign that either management is making decisions on factors beyond just quality...or that their methods for identifying quality are not sufficiently well developed.
introducing them with the knob turned full up. Listen to the screams, then gradually bring the knob back until the normal level of screaming has returned. This appears to be how they judge the right place for the knob in each major algo change.
I think G is just trying to hard. Ink looks fine to me and it is worrying if the sites on G and ink are completely different.
I'm also seeing a trend of people trying to dump their sites for dollars.
I'm not Brandy was even real (may have been a continuation of a fix of Florida). But, in any case, there were some major tweaks that hurt some of my sites and helped others. I can't figure it out!
Four of the sites I run are all holiday destination sites, after florida we jumped up the ranking considerably on all sites, but Brandy and Austin returned them to pre-florida.
Looking at the returns for keyword destination, all (and I do mean all) of the top ten sites for each destination are hosted in those locations (ie. that european country). I don't know if this means anything, but my sites are not hosted in the locations I am trying to break into. I will be moving them very shortly.
i dont see that at all...maybe that would apply to smaller niche sites who would be best placed to build a useful site on the topic...in larger searches i see the big linking sites dominate...with their subs taking several city rankings....
Not only did the 64 results not move over to the other datcenters in the few days that he mentioned, they NEVER moved over. They are just gone.
steveb
That may be true for the search terms in your field, but not mine. The results from 64 moved over to all the datacenters for the terms in my field.
wellzy
Looking at the returns for keyword destination, all (and I do mean all) of the top ten sites for each destination are hosted in those locations (ie. that european country).
I rank #1 for keyword continent, #2 for contintent keyword, #2 for city keyword, #7 for country keyword, #7 for keyword country, #6 for anothercountry keyword, despite the fact that my hosting service is in a different city, country, and continent.
I don't see any eveidence of the 64 results moving anywhere on the Internet though. The 216 results swallowed up the 64 ones like a minnow. A few trivial aspects of 64 survived, but on the whole 64 was simply never used.
And just for the record, I happened to rank a bit worse across the board on 64 compared to what was there before, so its not like I'm sitting here wanting something that would have been good to me. Now neither of those exist, and we are left with not-very-spammy but much weaker results, with anchor text-based pseudo-sites riding high.
Perhaps its the hosting , although how many global sites are hosted in the US? Perhaps its the topical content that specifically mentions the region more. This whole concept of topical sites seems to be throwing things on their head , and hence the floor is wide open for speculation .. which I agree with incidentally ..but nothing must be taken as gospel untill enough annecdotal evidence is gathered... how about shoving your worst performer onto a server in the region?
Not what I'm seeing. I'm seeing much better relevant results now, with a few spammy pages sprinkled in. Florida had no spam, but also terrible relevancy. Let's face it, broad matching and/or stemming produced the worst results in Google's history. I don't think it's possible to have highly relevant results without some spam. I wish that were the case, but it hasn't happened yet.
For me, the results now are littered with sites that appear to be on topic but are fully useless... basically well-seo'ed air.
Suppose the general topic is baseball. Someone throws up a piffle site that lists all the major league baseball teams, with one page devoted to each team, where they list team rosters from last year. Add a site map, and 60+ pages of links and you got a 100 page site. Call it keyword-baseball-teams.com. On each page link to a parent generic sports retailer.
Such a site adds no value to the results whatsoever. The information can be found on literally thousands of other sites. Now suppose this site is part of a family with tens of thousands of (similar) pages, which will give it boatloads of anchor text, which will get the Chicago Cubs page ranked in the top five for "Chicago Cubs" searches... even though the "content" is nothing but an outdated roster and a link to a related retailer.
This is what is doing well this week... and this is NOT a very "relevant" result. Yes, its on the topic, but it's junk. It's even worse than results that would have ranked in the top five an article in the Chicago Tribune about the Cubs. Given the choice, I'll take the "bad" result of the original content Tribune article over a totally worthless, lightweight bit of nothing.
"Relevant" are sites that have actual unique information about the Chicago Cubs... whether that be articles about their history or selling memorabilia or photos or any of several other things.
Real content is relevant. These results are full of PRETEND RELEVANCE. Pseudo-relevance. Fake relevence. Sites pretending to be on the topic but actually absent anything of merit except links/leads to the true point of the site (in the above example, the generic sports retailer with hundreds of thousands of pages).
Nope, these results are far less relevant. I care if a site or a page actually IS about something, not what it is PRETENDING to be about.
I think we all do. The problem was, high authority sites that had NOTHING to do with the search phrase were in the top SERP's. They just mentioned the words in the search phrase on their page somewhere. That is my definition of non-relevant results.
Now you get pretenders along with the good sites. At least that's what I'm seeing for the many searches I'm doing.
