Sorry this site requires JavaScript to be enabled in your browser. See the following guide on How to enable JavaScript in Internet Explorer, Netscape, Firefox and Safari. Alternatively you may be blocking JavaScript with an advert-related or developer plugin. Please check your browser plugins.

An exploration of PPR and the possible impacts it could have on search. With behavior metrics and personalized search becoming a popular trend among SEO’s we might be best served by understanding more about how search engines are going about such systems.
Comments5 Comments  

Comments

Avatar
from theGypsy 1914 Days ago #
Votes: 0

he he...thanks for the Sphinn Moving Man.... as we suspected, very little chatter. It will likely take another 10 months before it gets play hee hee... It’s good to have this out there at least for referrence once we dig a little deeper - it really is a far more viable area to look at than straight implicit signals.

Avatar
from jmaulson 1914 Days ago #
Votes: 0

so David, what does it really mean? at what level personalized factors could come to play? just before the results are displayed, as the last filter? or as the first one (like telling G what sites it should search through)? ’personalized pagerank’ is a contradictory term as far as I understand it: it is either ’popularity contest’ (PR) or ’personal preferences’ ("BS reinforcement"), there can’t be any synchronism here. The ’temporal factor’ you’re mentioning is of most importance here: should I be able to find new information or be denied it by my past searches and recorded preferences? And let’s not forget about G’s devotion to diversity on SERPs: at what circumstances factoring in personalization (localization, temporalization) becomes just a zero sum game? Does it ever? 

Avatar
from theGypsy 1914 Days ago #
Votes: 0

Ok, the first thing to remember is that personalization isn’t an upheaval in the results set. It is maybe 1-3 added URLs to the top 10 (usually only 1-2). It is not about 2 different people seeing 10 dfferent results. That it simply not the case (we published a study last year on this).As I touched on in the post, a lot of these types of calcuations aren’t truly personalized on a granular level, but more groupings of related users based on search history and possibly other demographic and behavioral metrics. So it’s not the doom and gloom of behavioral modifications and the death of rankings as some would have us believe. The concepts and the research don’t bear that out.

Avatar
from jmaulson 1913 Days ago #
Votes: 0

So could we just say that Google will simply look at more signals on both sides (instead of just to a query it will try to make the results relate to ’recorded behaviour’ and instead of relying mostly on links it will now start to pay more attention to statistical models of users, personas)? Would this be "a real change (we could believe in)" or just anoher level of the old language/country relativization of SERP content? Should we start buying Billboard & Outdoor Display advertising to influence personal search history and associated clickstream? or just use Twitter, for the time being? Peter Da Vanzo has an interesting post on it http://sphinn.com/story/97998 

Avatar
from theGypsy 1912 Days ago #
Votes: 0

There are sooooo many problems facing search engineers when it comes to engagement/behavioral metrics. Click data and engagement metrics are noisy and the IR community is still struggling to find clean usable data. I left a comment on that post (on SEO Book)

Upcoming Conferences

Search Marketing ExpoSearch Engine Land produces SMX, the Search Marketing Expo conference series. SMX events deliver the most comprehensive educational and networking experiences - whether you're just starting in search marketing or you're a seasoned expert.



Join us at an upcoming SMX event: