Forum Moderators: Robert Charlton & goodroi
Rank_1:1:6 Rank_1:1:5 Rank_1:1:4 Rank_1:1:4 Rank_1:1:5 Rank_1:1:6 Rank_1:1:5 Rank_1:1:0 Rank_1:1:1 Rank_1:1:5 Rank_1:1:2 Rank_1:1:3 Rank_1:1:0 Rank_1:1:0 Rank_1:1:3
These figures are for a site that has a current toolbar of PR4.
I searched the web for an explanation, but failed to get a good one.
Is there any idea as to what all these magical numbers mean?
I would not assume this though - not all sites have the homepage as the top result.
exactly same for my site, my home page is nowhere in the list in xml file
I also tried looking at my cookies but i see no google cookie besides the utm codes for analytics..
Can someone post a link to the tool to view your checksum? I can't find it via a websearch.
go to iweb and use pr prediction tool
The result is broken down into URL_1, URL_2 etc. of the page you're checking and all the "related" pages, showing URL, PR, summary etc. It appears to show the actual snippets of the pages it uses to judge the theme and relevance of your page*, i.e. anchor text and text located near your URL on the referring page.
*on further inspection the snippets don't look all that relevant, and don't take into consideration the search term. Still, may be an indication of how Goog sees your overall theme?
Not sure why it picks these particular 9 other pages for a sample though. They all appear to link back to the original URL but bear no resemblance to the Google "related" or backlink queries. May actually be the real deal.
This is a gold mine! Excuse me while I ahem research some top-ranking competitors.
Filter:0:
for all my sites when using the =filter
I thought it would actually be the 'Adult Filter' flag, so I did a check on a few domains that would be hidden due to such filter but alas; they all say Filter:0: on all domains I check.
Has anyone found a URL showing a different result?
>>>>Not sure why it picks these particular 9 other pages for a sample though.
They are the results for the search - whatever that search maybe.
Eg - if your query for the xml page is www.domain.com - do a search on the same dc as the xml query for "www.domain.com" - you will see the same results - the tool mentioned does not have to be used just for domain names - and search will pull up the results (except cant get spaces to work)
Also the snippet/title etc in the xml matches the snippet in the search results.
In an earlier message, I wrote:
I noticed that those figures correspond to the current PRs of other pages [inbound links]that are the subsample of backlinks to the site.
Today, I tried to study the <RK> values from another perspective: for sites filtered/penalized.
During late 2005 and early 2006, I was actively interested in some spamming sites. One of them was performing extremely well in Google. I wrote and published a short article describing the technique this site utilized to manipulate the SERPs. Google was quick to act stripping off its PR (that was 6 at that time), and suppressing it from SERPs.
Today, I queried its PR along with PRs of a couple of other banned sites.
After checking every link in xmls, I noticed that the <RK> values do not refer to the PR of the page in question.
I guess that the <RK> may be the PR <b>inherited</b> from the backlink in question.
Banned Site 1:
Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0
One of the links I checked has a toolbar PR3, and a current PR2.
Another link has a toolbar PR3, and a current PR3.
Banned Site 2:
Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0
One of the backlinks has a toolbar PR4, and a current PR5 (increased!)(<RK> = 0)
Another backlink has a toolbar PR3 and current PR3
A third backlink has a toolbar PR0 and current PR4 (jump!) (<RK>=0)
A fourth backlink has a TB PR0 and a current PR3
and so on..
These observations suggest that the <RK> values may be the inherited PRs rather than the PRs of the linking pages.
No - you are still thinking that the xml page refers to the domain - and I assume you are saying that the <rk> values being displayed for the domain you are querying is the PR being passed by the sites to that domain on the search you have performed?
EG. Are you saying that for domain www.example.com the values:-
Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 Rank_1:1:0 - are the PR values passed by each page to the domain www.example.com?
If so this is wrong - the xml query just relates to the search page for the query - not for the domain. Dont think of it as a domain query - just a search query.
Eg:-
[72.14.207.104...]
is just this page in xml format
[72.14.207.104...]
The query is not relating to the domain but to the search - so there is no way that the <rk> values can be said to relate to a passed PR value.
>>>>After checking every link in xmls, I noticed that the <RK> values do not refer to the PR of the page in question.
Yes, because PR is out of date - these <rk> values may be current, more current.
<snip>
[edited by: lawman at 11:22 am (utc) on Feb. 17, 2006]
[edit reason] No URL Drops Per TOS #13 [/edit]
That is not what the tool does.
The tool looks for the www.example.com in the xml file and returns the <rk> value for that page - none of this add up the last digit and divide by the number of results.
Also the article argues that as the <rk> values do not match the current PR then the tool is wrong - as PR is out of date then that covers that argument. We are talking about possible furture PR - not current
Poor article.
Shame the guy actually missed what the tool does :/ - as if he anaylsed properly he may have been on the right track.
The article explains it fully
The <rk> value is vague in the Google definition.
"RK Definition: Provides a general rating of the relevance of the search result"
The RK value does not change on different search results for a start - eg pulling up webmasterworld in a webmasterworld search or in a search for PubCon are Registered Trademarks will result in the same RK value.
[edited by: Dayo_UK at 1:42 pm (utc) on Feb. 16, 2006]
so if you search for "www.yoursite.com" the RK value is how relevant the search result - not to do with page rank
I edited my post sorry.
As above - the RK value does not change, even if the search term changes.
It is vague term and certainly does not change with regards to search term - it is a static value for the page.
The case for it relating to page rank is stronger than not - the article does nothing to change that fact.
What we seem to have is a static figure (although changes on dc) for a page that looks like it is a ranking figure - this figure is very very close to PR on a lot of occasions. (although PR is very out of date)
[edited by: Dayo_UK at 1:50 pm (utc) on Feb. 16, 2006]
Yes, you are right in that the xml relates to the search, and not to the domain.
I rechecked, and perplexed at first, since I got discrepant values. Then, I noticed that I am checking two different datacenters: 72... for "hello" query, 66.249.. for domain query. ): When checked with the same DC, identical values are obtained.
Thank you for correcting me.
a general rating of the relevance of the search result
If it has anything to do with PageRank, then that definition would be different and would at least mention it
Also as this <rk> figure does not change based on the search term we are left with the definition of:-
>>Text (Integer in the range 0-10)
>>"Provides a general rating of the relevance of the search result"
Which as we know is not dependent on the search term - if it is not PR it is something very close to it.
[66.249.93.99...]
And it returns results for that - surely blah cant be a domain name, and as a result, those numbers clearly cant have anything to do with page rank or future pageranks
The article is wrong. It claims the following:
There are three steps to how iWEBTOOL predicts your future “PageRank”1. It sends a query to Google with the URL you specify and Google returns a result which looks something like:
Rank_1:1:6 Rank_1:1:6 Rank_1:1:6 Rank_1:1:2 Rank_1:1:4 Rank_1:1:4 Rank_1:1:4 Rank_1:1:4 Rank_1:1:3 Rank_1:1:3 Rank_1:1:3 Rank_1:1:3 Rank_1:1:3 Rank_1:1:3 Rank_1:1:5
2. It takes the last digit for each 3 digit combination and adds them up
3. It then works out the average of these numbers and returns the result as your “predicted PageRank”
Hello/Blah/Anything ... is irrelevant if you cannot
find your own domain. Obviously blah is not a domain
so hence the <RK> data cannot be found for this. BUT,
the <RK> data CAN be found for other domains that are
related to the concept of blah.
Also, an integer between 1 and 10? Read the Google XML
specifications, and you will find there are NO other
variables of this type. Between 1 and 10? Sounds pretty
plausible to me ... and I am pretty sure Google would
NOT label this as pagerank (even it it is!)
[66.249.93.99...]
ht*p://www.blah.com/ has RK 6 - Current PR is 5
ht*p://www.blahdvd.com/blah/ has RK5 - Current PR is 4
ht*p://dailyblah.com/ has RK6 - Current PR is 5
ht*p://www.eblah.com/ has RK7 - Current PR is 7
etc
So they are pretty close - and certainly could be possible with a PR update - esp as Displayed PR is old.
However, one thing I have noticed is that <rk> figures do tend to be on the higher side than low side.
hmmm...check these out..did some digging -)you can replace =Rank with the following also
=Level
=Filter
=Hostname
=URL
=Results
=Summary
=Title
=CacheSize
=Link
I found the Summary command an interesting one. Embedded in that information is the location that I used when registering my domain. That address and telephone number appears nowhere on the domain itself (although I've been thinking about changing that), so this must be pulled from the registry information.
Has anyone tried to click on both current PR and predicted on iwebtool, and then compare the results?
