| 7:40 am on Feb 28, 2004 (gmt 0)|
|Won't we now have to optimize for both (as well as Google)? |
Michael - I don't see that they're mutually exclusive. In my experience, sites with good "classical" on page optimization and good linking perform well on all three algos. If anything, the new Yahoo algo is a little closer to Google than was the Inktomi algo.
This is apart from the Florida/Austin changes on Google, but I must say that some really irrelevant sites that I saw doing well for certain searches on Florida/Austin... sites which jumped out like a sore thumb as not belonging, though you can see why they're there... also did well on Inktomi, and continue to do well on Yahoo, which maybe suggests even more similarity.
| 7:57 am on Feb 28, 2004 (gmt 0)|
"which maybe suggests even more similarity" = "quality and quanity of links matter to both"
| 8:43 am on Feb 28, 2004 (gmt 0)|
|"which maybe suggests even more similarity" = "quality and quanity of links matter to both" |
steveb - I agree with you that this is what the algos have in common. This is sometimes also what fools them, at least partially.
In the "sites which jumped out like a sore thumb as not belonging," there's something else going on, and it's just an odd combination of factors that approximate the directory effect we were seeing in Florida/Austin.
In one case, eg,in a search for keyword widgets placename, the PR6 homepage of a web design company that is located in PlaceName ranks where it's not really relevant.
The page features a link to one client with anchor text containing "keyword gizmos," with "keyword widgets" in the description under the link... and another to a client with "widgets" in the anchor text.
Several of the designer's other clients are in "PlaceName" and so identified on the page... and my guess is that the page has some inbounds with "placename" in the anchor text. It's not really a relevant page, but the combination throws all of the algos, and you can see why.
If the algos paid more attention to the title tag on this page, it wouldn't rank; but I'm guessing there are other factors as I mention above, that, at PR6, override the title tag. I haven't analyzed the inbound links.
| 8:52 am on Feb 28, 2004 (gmt 0)|
I'd add that while PR is a Google thingee, the smae basic concept is at play with Yahoo... which is again the same translation as above, quality and quantity of links count for both search engines. Of course, a lot of other things do too, but they have this in common to a significant degree.
| 1:15 am on Mar 1, 2004 (gmt 0)|
Personally, I don't think Yahoo optimization = Ink optimization. The databases are different, the SERP's are different.
The way I understand the new Yahoo search is that it is a hybrid, comprised of the different parts of each of the engines that they bought.
I'm trying to see Yahoo as a whole new search engine that I have to understand. By treating it with this view, I'm not carrying any "baggage" into my analysis and looking at it from scratch, in a way that will help me understand the new algo.
| 8:13 am on Mar 1, 2004 (gmt 0)|
|I'm trying to see Yahoo as a whole new search engine that I have to understand. By treating it with this view, I'm not carrying any "baggage" into my analysis and looking at it from scratch, in a way that will help me understand the new algo. |
2_much - I think this is probably better advice than my oversimplified comment in my earlier post. I think the differences show up on phrases that are more competitive, and/or on sites that have lower Google PageRank and linking that's not as targeted.
On my sites... and this is based on not very extensive testing, but on benchmark comparisons that have provided a good general guide to me... those with higher PageRank do surprisingly much the same on searches across the three algos (using MSN advanced search as a replacement for PositionTech Pure Search, which is now gone). Ditto with not highly competitive searches, even with PR5 sites.
But start getting into very highly competitive two word phrases (the Google "money phrases") with PR5, and I see that the results start looking way different.
All that said, with good onpage optimization, I would bet that more and better links would bring my results in Yahoo and Google back into alignment. Inktomi looks to be easier to beat... It definitely looks at targeted links, but appears to be less reliant on them. I'm sure that freshness of the indexes is also a factor, and I haven't tested this at all.
Yahoo also has some big dupe-page problems that it hasn't sorted out.
| 1:12 pm on Mar 1, 2004 (gmt 0)|
|The databases are different, the SERP's are different. |
... and the algorithm used to produce the SERPs are different (right?).
Thanks for the advice and analysis guys.
Would I be correct in summing it up this way: the basic language of SEO is the same for each of the major engines now, but individual dialects exist?
| 12:44 am on Mar 5, 2004 (gmt 0)|
Although I understand and agree with the arguments presented, I think it it worth pointing out that although Ink optimisation does not currently equal yahoo optimisation there is some correlation between ink results and msn.
That is, although it is useful general advice to optimise for all earch engines, the current results do seem to indicate that MSN is using a sorted version of inktomi, in my area anyway. So ... Ink = Yahoo is false, but Ink = MSN might be true.
| 1:36 am on Mar 5, 2004 (gmt 0)|
I haven't had to change anything - yet ;)
Everything doing well on old INK doing just dandy on the new Yahoo! with an extra boost (IMO) from having a directory listing!