| 12:30 am on Sep 22, 2006 (gmt 0)|
I think it might be because
of different datacenters being used, plus your nearest data centre is probably diff too
plus, the SERPs are dynamic, shifting sands you see, as the intrepid SEO's battle valiantly to be numero uno
| 7:34 pm on Sep 25, 2006 (gmt 0)|
In other words - the "average" doesn't mean a thing...
| 9:17 pm on Sep 25, 2006 (gmt 0)|
average mean't something at the datacentre, and at the instant in time that the computation was made,,
but , at a different data centre, an different point in time, where, and when you finally access the SERPS, the results inevitably must be different
| 10:27 pm on Sep 25, 2006 (gmt 0)|
I know the technobabble...but as a tool for working SERP, the average provided by Google is worthless because it "might be" any one thing on any one datacenter, and provides no average that can be measured, verified, or quantified in any manner.
If I'm wrong, then I'm willing to be educated, but the question is about the utility of the tool, not how the average is arrived at.
BTW - while our positioning has been all over the place on multiple datacenters, it has never once been above what is called the "average" over the last 3 weeks.
| 10:49 pm on Sep 25, 2006 (gmt 0)|
"Technobabble" thanks Mjunge,
One of the things I find about google is that they are relentlessly precise , this results in a lot of work for webmasters to master all the variables,
Webmasters tools is a worthy attempt to provide a snapshot picture, of your websites standings ,
you cannot freeze time, nor can google really be expected to give up a realtime picture of our Serps position,
a weekly update is quite decent an I hope yahoo & msn provide something similar soon
| 2:50 pm on Sep 26, 2006 (gmt 0)|
I agree with everything you said...but...
Over the last three weeks none of our averages have changed - for over a dozen search strings. This doesn't make sense with a constantly/weekly updating Google.
Over the past few months as I've worked through the threads on datacenters I seen folks obsessed with they "why" of differing results and the location, hops, loading, and so on of the different datacenters. What I haven't seen is a coherent discussion of which datacenter, or colelction of datacenters, you should use to determine where you rank out and how you compare...hence my use of "technobabble".
I can only assume (since there's only two of us here) that you have not had the same experience. That leaves two options: 1) that there is something wrong with the tool as it relates to my website/search strings or 2) that there is something wrong with my website as it relates to this tool. I'm looking for other opinions of the tool to try and narrow down which of the two issues it might be.
| 11:01 pm on Sep 26, 2006 (gmt 0)|
Over the past 4 weeks it has updated 2 times, one off the google reps suggested it would be a weekly event, but it hasn't quite worked out that way,
One hopes that they read this an go push that update button again, an again, every week
And I find it quite usefull , sometimes , my current SERPs position is within 5 positions of the stats figure for one or two keywords,
| 10:41 am on Sep 27, 2006 (gmt 0)|
I think we might be cross talking.
Our rankings still change daily...likely due to the datacenter idiosyncracies. But that is not what the original question pertains to. What I am talking about is the "Average Top Position" in the Webmaster Tools (Statistics/Query Stats) section of Google.
If the "average" that I see there is higher than any of the rankings I have seen when surfing or checking the datacenters, how is it an average? What is it really an average of? Or, what could I have wrong on my website that is providing an invalid result.
| 12:19 pm on Sep 27, 2006 (gmt 0)|
Nope, I understand you, and the change I was refering to pertains to the statistics in webmaster tools, these have been updated 2x in 4 weeks
I am also curious as to why more people don't do threads about these statistics
But I think its because most people are used to google not updating the public figures,
eg toolbar pagerank, links etc etc
For the webmaster statistics to be really usefull, they should be more regular, like clockwork, like every Monday 9 am GMT or USA time whatever :-)
| 4:17 pm on Sep 27, 2006 (gmt 0)|
agreed...if they advertise (which they do) a running average over the last three weeks, then it does need to reflect reality, not an outdated snapshot.
I just wish my average was what they report my average to be!
| 10:08 pm on Sep 27, 2006 (gmt 0)|
We follow google sitemaps very closely for our news site. Those search query tools they supply are highly inaccurate. There is one term that said we were ranked #6 for. If that were true, even for a week, we would be very rich right now. There is no way they are correct. For some of the keywords they are correct for, the results are from weeks ago, not from different datacenters, but from different centuries.
| 6:30 pm on Sep 28, 2006 (gmt 0)|
Coincidentally, I have been working on a blog post over the last week about this very topic. I just posted it this morning and I think it might help answer some of these questions.
Let me know if you have more questions and I'll do my best to answer them.
| 10:11 pm on Sep 28, 2006 (gmt 0)|
Why aren't the statistics being refreshed weekly, as expected
Okay I haven't read the blog yet,
| 10:15 pm on Sep 28, 2006 (gmt 0)|
Okay, I just read the blog
Every Monday you say, well do you mean from next Monday, this will be the case?
| 10:36 pm on Sep 28, 2006 (gmt 0)|
Yep, should be next Monday. We decided to go with Mondays regularly so that you would have an expected day to look for changes.
| 12:10 pm on Oct 10, 2006 (gmt 0)|
Its Tuesday and I do believe that the big red statistics refresh button hasn't been pushed yet
Our weekly Monday fresh statistics fix hasn't come in
| 11:01 am on Oct 11, 2006 (gmt 0)|
Wednesday, the statistics are now archaic :-)
| 3:36 pm on Oct 11, 2006 (gmt 0)|
I can't say when these stats are refreshed, but I can say the page they list this month and last month as the one having the highest page rank makes no sense at all.
It has few incoming links from other sites, is several clicks away from the home page, and there's nothing special about it that would make it stand out in any way.
The PR in Google Toolbar shows a 2 for that page, with many others in the site showing 3 or 4, and the home page showing 4 or 5, depending on datacenter, and sometimes 6. It makes no sense to me at all.
If the rest of the stats are as reliable, I don't trust any of them.