Forum Moderators: Robert Charlton & goodroi
its not the trademarked ghost dataset that went missing, and it wasn't a rebuild like the halloween update.
No, but the overall technique has a familiar feel to it. More than one dataset may be involved this time - and perhaps many more. Interesting that three weeks ago we were hearing reports of googlebot spidering like crazy, and in recent days, reports of googlebot not even showing up for some sites.
[edited by: tedster at 5:09 am (utc) on July 15, 2009]
ADDED-
whitenight, my observations suggest the slow-fold means total resolution will take longer than you anticipated (not to undermine the fact you correctly called the end-game). I mention it because alot of people may panic if their rankings do not return as per your Tuesday schedule, which is now being taken as read. I'm thinking Wednesday or even Thursday before it completes. Agree?
Could be. But remember, when they start the full roll-in, it happens fast...
Note: some people are going to panic no matter what happens, it's an annoying fact of the human mind.
its not the trademarked ghost dataset that went missing, and it wasn't a rebuild like the halloween update.
What Tedster said. =P
Predicions require critical thought, not skilled rhetoric.
lol don't like my skilled rhetoric?
Sorry, but my posts on this board are mostly rhetoric.
It has it's purposes beyond predicting updates or explaining SERPs.
ever wanna change the world? Once you figure out how to do it using critical thought, rhetoric is what gets people to accept your ideals
Im practicing going to Bing everyday... and YOU should too...lets ALL start finding reasons to use BING. God-gle is only as BIG AS WE HAVE MADE THEM.
Now this is rhetoric I can always stand behind. :)
[edited by: whitenight at 9:06 am (utc) on July 14, 2009]
Its others who are satisfied when they think they understand something, and grasp to a substandard explanation for that comfort. Its better to know what you thought at the time, rather then be revisionist and think you had the same explanation all along.
Also, the official Google PR campaign will start after the event, and it helps to have alternative versions of events in the public domain beforehand.
Anyway, interspersed with your agenda-pushing, are often good datapoints, better refutations of weak data, and accidental spills of actionable information.
And, scarily, it might just be that someone else has some analysis to push
Slightly aggressive tone not a sign of disrespect- I know you too are a fan of more robust converstion than is normal on these boards
Also, the official Google PR campaign will start after the event, and it helps to have alternative versions of events in the public domain beforehand.
Yea, i've slowly given up on this policy here.
Google employee #1-#4 comes along saying something, and it's nearly impossible to convince certain people otherwise.
I like your analysis so far, Shaddows.
Lack of open agreement on my part can't be construed as disagreement.
This is truly a "real" update...
As someone said earlier, more along the lines of Big Daddy or Florida in terms of what the "overall goals/changes" are...
So I would look at what tedster pointed out early on, as to what Goog is ultimately working on.
I'll add this little tidbit and one can do with it as it will.
Goog may be trying to TRULY get the best ranking pages for a term, instead of the defacto "site authority" for terms.
This is more along the lines of what you've noticed as well.
Now, after it all settled down, it became clear that google weren't intentionally tweaking the serps, it was a side effect of what was being put in place to allow everflux, this marked the end of the monthly google dance and the start of a rolling update.
My point, what we are seeing now might well be linked to some other infrastructure/crawling changes that google are putting in place.
It would explain, why, like in 2003, there wasn't any real pattern to what was happening.
Goog may be trying to TRULY get the best ranking pages for a term, instead of the defacto "site authority" for terms.
I could go along with a complete structural revision of the architecture of the algo. It would be more obvious on the competitive terms due to the level of inherent manipulation, plus the sheer number of permutations available from the high link-count contributing to competitive success. Not to mention the number of SEO eyeballs watching for any movement.
Any pattern would be hard to discern from the actual SERPs, although processes and timings would appear similar to November rebuild.
My only problem with this version of events is the sheer length of time it has taken to resolve into usable SERPs.
Still, the roll-up will be worth seeing.
NOTE: I wrote the above a while back. I note UK_Web_Guy has a similar take, which I can't disagree with. So, Big Daddy II then?
Here's a stab in the dark:
You have an algo, you spot deviant behaviour, you write either a separate routine as a patch, you carry on.
At some point, you have too many patches, and need to incorporate them all into the algo proper, possibly with a mechanism that allows you to add new routines straight into the core algorithm.
You put it live (with some false data distortion which you can easily remove internally to hide the process from public analysis) and watch the results. Its not quite what you expect. You leave it live and start on-the-fly bug-fixing. When its behaving properly, you start removing the data-distortion, and allow the results to normalise.
Again, it superficially resembles previous updates, but protects against actual analysis of what happened.
Sorry, but my posts on this board are mostly rhetoric.
Aren't everyone's?
Goog may be trying to TRULY get the best ranking pages for a term, instead of the defacto "site authority" for terms.
Not what I'm seeing on Google.co.uk - the opposite is true for checks I've made the past couple of days.
and the wiki monster would be hurt too.
If trust-rank was completely removed from the algorithm Wikipedia would still rank high. The sheer number of links and the internal structure give it a huge advantage.
On top of the changes that are expected:
- A reliable source indicated to me (she slipped, put it that way) B*ng (MSoft) hired at least a couple of senior former G* engineers early 2009, they where part of the Algo crack team at the Plex.... (make your own analysis and draw your own conclusions here, but what happened gave G* more reasons for even larger Algo "re-structuring")
- A surprising corporate synergy will be announced just before the end of September from G*, the expected changes are directly related to that.
Disclaimer: In no way this post should be taken in all seriousness as a "tip" for financial speculation, having said that, only time can tell. In other common English words, Don't bank on it!
Bookmark this post, and you'll see what I meant!
this will not properly settle until the end of September.
lol, how's that for panic?
Seriously, what's your definition of "properly settle"?
As much as I'd like to see the SERPs this crappy all summer for the sake of BING's marketshare,
for the exact same reasons,
i don't see the SERPs being this crappy for another 2 1/2 months...
Even if that means Goog rolls out a "stop gap" algo in the meantime, they simple can't afford to let this go on too much longer.
So back to my original question,
"What's your definition of 'properly settle'"?
So back to my original question,
"What's your definition of 'properly settle'"?
It'll get even worse, and apparently it's unavoidable, those "change-and-test" and "rock-and-roll" have to be done live and in real time. They have few hundred "reviewers" why can't those act as test-beds within an intranet-like small scale simulated environment, no we have to put up with it live!
This is a major year for G*, it's not a make or break, it's the "stay or break" year. In a way, I am glad those companies are driving them to this, yes MSoft is one of them and credit to MS, they are not hiding their tactics, but others are and G* is in a beat them or join them dilemma. What defined and still defines G* is their algo and the people who decided to leave teoma/ask, alta vista, yahoo and northern light bringing in the ratatouille recipe (most of the current algo used today) with them. Other services it provides were not G*'s typical offerings, but unlike yahoo which got it the other way round, G* hit it on the head: The search offering as the core, other services as add-ons, with that formulae they achieved a phenomenal IPO success.
You can treat what I say here as speculative statements and that will be OK by me. Some conclusions are indeed drawn from observations like many here and some are from purely speculative analysis. All what I can say and expect to happen is, the search industry will soon see a de-monopolization and that should be good for users NOT for G*. In the 1990s it was that way to some extent, but the technology was not that good and more often than not we did not have a favorite search engine, we had 3, 4 or even 5, hopping from one to another trying to find what we are looking for, many ISPs spoon fed members with their own branded in-house solution and some even in the early stages forced it through their branded browser and nothing else. I remember having a hard time trying to use Mosaic and later IE while a Compuserve or AOL account holder. The difference is, more than one search engine for a change will be highly "relevant" this time, take your pick is the way to go and that will split market share big time.
Today, I discovered one new link to one of my inner pages from an inner page of what appears to be an "attack site". If I Google the domain name, this site comes up in the serps with the "this site can be harmful to your computer" blurb. I have nothing to do with this site. I found the link from them in Webmaster tools.
Could this link to me be hurting me in the serps and if so what is the best remedy? Delete my page? It is not a high traffic page.. could remove it and also remove it from the index...put up another one renamed?
What is the best course of action if this is indeed the problem?
It has been in the top 10 for the past 2 days.
I have never seen it stick around once things settle down.
I'm noticing that rankings are improving dramatically when I link out to an authority on the subject (or an authority page on my topic).
Yes, despite all the FUD in the thread, this does have all the makings of a fairly standard update. Three data sets merging and the spam thinning... hopefully the overabundance of index pages will start getting replaced more by topical subpages, and then in awhile, poof, back to a stable index. It has been a long time since major changes, so it shouldn't be a surprise that this goes on for awhile.
Yes, same here. From where I am sitting it seems we are close to the end of this update.
Any one noticed this?
I have 3 website of which 1 has high levels of visiotrs through and also has ppc activity.
The other 2 do not.
the traffic website has good positions and held them while the others have dropped and have higher banded websites ranking above
You'd need to check referal logs to understand where your own traffic was coming from. But as there is a mighty strong correlation between traffic and ranking, I suggest that things have not changed much for the niche you're monitoring.
(High Ranking -> Traffic NOT Traffic -> High Ranking)
my site that had been back to normal on sunday & monday has now gone again!
im sure it will come back again soon HOPEFULLY for the last time!
I would delete the link and look for a possible hack as the link got there somehow to direct users to a hacked site and quite odd it is one of your high traffic pages.
Delete the link search for a possible hack and send resubmit the site with what you did to fix the issue.
I have noticed that four sites in my niche that normally rank mediocrely have bolted up the results, and it has been their month for their seasonal/yearly traffic peaks (moreso than the other sites in the niche).
On the other hand, all four have been relatively under-appreciated by Google in the past, so perhaps the current changes just reflect a better appreciation of the assets these type of sites have compared to the rest of the universe.
There is no doubt though that this is the one development in my niche that suggests something new in the secret sauce.
