Welcome to WebmasterWorld Guest from 126.96.36.199
Forum Moderators: open
I'm surprised that lack of the third term in the page title didn't prevent this. It looks like Fast has dialed up link popularity... but at the expense of onpage elements.
On Google, link content/context and other factors keep this from happening, at least to such an extent. The page does occasionally rank for terms where I'd prefer the (more general) home page would rank, but the terms are usually prominent on the page and/or in inbound links.
Heini - This may be more than you ever wanted to know...
The phrase without the pulldown menu modifier is extremely competitive. The menu modifiers are abbreviations of location names, and the search takes the form <location keyword1 keyword2> (all the words).
For either an exact or all-the-words search on Google for "keyword1 keyword2," the site's in the top 10... with the heavily linked subpage coming up along with the index page. Competition is 5.6-million all-the-words on Google, 360K as an exact match.
Fast shows 64-million for the all-the-words search, for which the site doesn't rank, and 1.8-mill for the exact phrase, where the heavily linked subpage shows up as #15.
When you add the "location" modifier, it depends how you do it. There aren't many 3-word exact matches at all. Numbers for other forms of searching the phrase are:
For all-the words:
<location keyword1 keyword2>
- Google - 1-mill
- Fast - 2.6-mill
This search returns the heavily linked subpage as #1 on Fast... Some relevant subpages that use the abbreviation start showing up at about #16 in Google.
For exact on the main phrase only:
<location "keyword1 keyword2">
- Google - 55K
- Fast - 103K
This search also returns the heavily linked subpage as #1 on Fast... Some relevant subpages that use the abbreviation start showing up at about #15 in Google.
Looking at the above, I could suppose that Google gives less weight to pulldown text. I'm not sure whether this is true or not. I did a very quick all-the-words test for the same location, but not abbreviated, to see the difference for analogous terms without the possible pulldown factor.
The relevant location page plus the home page come up as #4 and 5 for this search in Google (2.3-mill pages). In Fast, the heavily linked subpage comes up as #16 (6.8-mill pages), and the relevant location page doesn't appear at all.
Another factor possibly involved here is the way the algo deals with 3 or more word queries.
Example: 3 word phrase, one hyper competitive, one still highly competivie, in combination hard to rank on top. The third very specific, not competitive. The page ranking top has the third word only in links text, it's the index page. The page dedicated to the three word phrase doesn't rank well.
The index page coming #1 also is #1 for the highly competitive 2 word phrase.
We have seen Fast introducing the automatic rewriting of queries, where 2 words where automatically rewritten to an exact match.
In this case this does not happen, adding quote brings different results. Still I wonder if somehow the two words have prefernce in ranking with the third given less importance.
>>We have seen Fast introducing the automatic rewriting of queries, where 2 words where automatically rewritten to an exact match.<<
Yes... and I think when they rewrite the words to an exact match, they must also give them a different a different weighting. An example I've cited before, of an all-the-words search for san francisco widgets, gives very different results when they rewrite san francisco to "san francisco". Since in the search universe the query addresses, "san francisco" is always grouped as an exact phrase anyway, the difference would have to be in the weighting given the phrase.
I was actually surprised, in the test search I just described, that the results on Fast were the same for both <location keyword1 keyword2> and <location "keyword1 keyword2">. Maybe, if the grouping had been Fast's grouping, not mine, they would have weighted the grouped phrase differently enough to change the results... or maybe this site just happens to do well on that phrase grouped or ungrouped.