|allinanchor: In Search of Greater Understanding|
A few unanswered questions...
| 3:46 pm on Jun 5, 2003 (gmt 0)|
OK I've read and re-read this string: [webmasterworld.com...] ... and other resources also ... but still do not feel that I have a complete understanding of the "allinanchor" command.
So many times I've seen groups of people act on assumptions that they *thought* were correct, but that later proved false or incomplete; I'm thinking that this might be one of those.
Specifically, I'm curious how the SERP's are formed for this command. For example:
--Is PR figured in - i.e., does identical anchor text from two pages with different PR carry different weight (I'm assuming yes, but is that certain)?
--Do two backlinks from the same external page "A" carry double the weight of just one backlink from that same page?
--On a two keyword search (keywordA, keyworkB), does backlink anchor text score more highly if it contains *only* those two keywords, versus another text link that *includes* those keywords but also contains "keywordC" and "keywordD".
--And my biggest question: are these SERP's we're looking at adjusted by spam filters, or are we seeing relatively straightworward results (if there is such a thing), prior to any spam adjustments?
I *think* I know the answers to most of these questions, but I'm not *certain* that I know the answers to any of them. Any help?
| 4:16 pm on Jun 5, 2003 (gmt 0)|
Now that I read your questions, I'm not sure that all of my assumptions were correct either, but I am under the impression that the allinanchor search results do reflect spam filters.
Right now across the nine data centers, I see different results for an important two word search phrase in my main category, and I'm assuming it is related to the spam filters.
Not sure however what other elements are factored into allinanchor results. I always assumed PR was in there, but I guess I could be wrong about that.
| 2:23 am on Jun 6, 2003 (gmt 0)|
Caveman, that's what I thought too - that the spam filters were in there, but now I'm not so sure how these are sorted.
Anyone else? I'm feeling really lame about this one...any help greatly appreciated.
| 9:32 am on Jun 6, 2003 (gmt 0)|
Interesting questions. Unfortunately, probably you won't get a definite answer.
My observation is that other elements are factored into the results. For me it seems that PR is part of the results. However, I think that only the PR of the resulting pages are taken into account. (PR of the pages with the backlinks plays only a role for the PR calculation.) Thus identical anchor text from two pages with different PR carry the same weight.
| 12:35 pm on Jun 6, 2003 (gmt 0)|
Thanks. I think that some other factors must be playing a role. Just like Caveman, we see different results across different data centers, so *if* the index being used at all nine data centers is the same, differing results *must* be caused by other factors.
What staring to bug me more and more is that if a lot of other factors are at play in these results, I question how useful the results are.
For example, if these results are modified by spam filters and show our site at #8 for a given keyword phrase, how do we know that without the filters, our site wouldn't be at #1? Without knowing that, it's then hard to tell if we've overdone it already, or haven't done enough. Make sense?