PR, Authority and Trust

[Chexone]

I've been reading a lot about the google algorithm and i know no one really knows except google, but it all kind of makes sense. Let's forget for a moment about the other 147~187 factors that affect SERP positioning, and focus on PR, Authority and Trust. So, is the following correct?

My page gets PR (link juice) from backlinks to it. The amount of juice my page gets from a backlink depends on the PR of the linking page, and the amount of outgoing links on that linking page (the given PR is spread evenly across all the links). Anchor text does not factor into the PR that is passed.

My domain/site gets Authority and hopefully a little Trust from quality backlinks that themselves have some of one or both.

Authority *may* be topic-specific, so if i've got a lot of backlinks with anchor text relating to "movies", i gain authority for movie-related terms from them. This explains the importance of anchor text.

Thus, if i've got lots of PR on my page (lots of backlinks) and my site has lots of authority for the search term "movies" (the anchor text of those backlinks and of backlinks to other pages on my site are related to movies) and by luck i've even gotten a bit of trust passed down from the American Film Institute linking to me, then i will rank high for a movie-related search.

Am i close?

But what if my site has authority, not only for "movies", but also for "books"? How does google know which page to rank high for a "book" search? On-page factors! Content!

Or does PR actually depend on the anchor text, and thus, a page can have Topic-specific PR?

There's something missing... Something like PageAuthority. Something topic-specific like Authority, but specific to one page like PR. No?

[goodroi]

hi chexone,

when reading up on the google algorithm you should be careful to take note about the time the research was written. the google algorithm has hundreds of changes every year and its over 10 years old.

several years ago google used to mostly use pagerank. now they look at trust, authority, usage data and many other things. the importance of pagerank has greatly devalued. personally i feel that google's use of usage data is much more important than many webmasters realize.

google has access to usage data from google toolbar, ip log files, twitter mentions and other sources. this helps google see how users really use the internet and it is harder for webmasters to manipulate this dataset. it is much easier for webmasters to manipulate backlinks and anchor text. that is why imnho google's pagerank is not as critical as it used to be and usage data is much more important than many people realize.

[Shaddows]

PageRank is not the story it once was, but don't forget about PR-like data. Trust, authority and vast areas of semantic 'logic' uses (or appears to use) the PR methodology.

Also, link juice is not split equally to links on a page. Prominance is important, and ROS is low-value (potentially negative, depending on use).

Topic-specific Authority tends to get allocated to page-groups (be it a whole domain, or a silo), and diffused relevancy strongly undermines this effect.

Trust is a bit different. I find it does not affect your actual rank, more your potential to rank.

Conjecture Alert---->Trust is a bit screwy at the moment- non-trust can rank high, which is being offset by UberTrust sites getting an artificial boost.

[tedster]

Trust is a bit screwy at the moment

The balance between near-to-real time results and trust is a pretty risque tango

[aristotle]

I wonder how the age of a site is related to its trust and authority. I've seen different opinions about its importance.

[Shaddows]

Age is a different thing. Stability of links comes with age, which might confer Trust, but there is no relationship beyond correlation.

Authority is unrelated to age. It is conferred from other authority sites, which appear to be hand-selected. You will need to do some homework here, but a good starting point (as it is for so many other things) is the end of October through November 2008 Monthly thread. That was the last time Google was properly caught with its trousers down.

[Chexone]

So all the voodoo magic has been replaced with intelligence? Great... ;)

So when you say usage data is important (and by the way, google knows way too much, with everybody using their free analytics, etc...), you mean, for example, that if a user clicks on a SERP listing and spends a lot of time on that site, maybe fires a few google conversion tracking pixels, and comes back the next day for more... google will know it's a good site and rank it higher?

[Shaddows]

More, if the right people tweet about it, and the right blog mentions it in the right way, and if facebook is liking it too, well you're quids in.

And quite possibly a marketing behemoth.

[aristotle]

Concerning usage data, one of the reasons Google created its Chrome browser was to be able to collect this kind of information. Examples of positive signals would be:

1. User bookmarks a page as a favorite

2. User returns to the page again later.

3. User saves a copy of the page onto own computer.

4. User visits other pages on the same site.

[Shaddows]

Going back to the OP, PR is about citations, which has been conflated with links.

MB has a thread going about it here [webmasterworld.com]

[Planet13]

The amount of juice my page gets from a backlink depends on the PR of the linking page, and the amount of outgoing links on that linking page (the given PR is spread evenly across all the links).

If that is so, then why do people say that it is better to get a link from within the content (say, in the middle of a paragraph), than in the footer of a page?

Maybe it is a misconception that "the given PR is spread evenly across all the links."

[Shaddows]

Its not "maybe" a misconception, its plain wrong.

To quote myself:

Also, link juice is not split equally to links on a page. Prominance is important, and ROS is low-value (potentially negative, depending on use).