A look into the origins of Google TrustRank and TrustBox

This article was originally written back in 2005.

I was helping a friend do some research on TrustRank and TrustBox to find some more information on the two topics. Here I will touch on the two.

First, the term trustrank first appeared in a research paper “Combating Web Spam with TrustRank“. To get straight to the goods, with their research, they are showing that they are able to successfully filter a major amount of spam from the web with a set of websites that have been manually reviewed. The research uses less than 200 sites as their trusted sources (the seed sites). As mentioned, the 200 sites are reviewed by an expert, this is used to essentially find pages that are deemed to be good, this part is done automatically as there are not enough resources to manually review every website.

In the research paper, section 6.4.1 has a very interesting comparison with PageRank Vs TrustRank. The following are quotes from the research paper

“PageRank algorithm does not incorporate any knowledge about the quality of a site, nor does it explicitly penalize badness.”

“TrustRank is meant to differentiate good and bad sites“

Second, we have TrustBox. One of the earliest mentions of Trustbox is from SEObook.com in which he has an article on Google Trustbox. Also Danny Sullivan gives us a great article on the evolution of links and the latest model is leading towards Trusted Links. Basically, the link evolution is broken up into 3 sections. The first being before Google PageRank, essentially, each link counted the same. The second phase, not all links are equal, some are worth more than others. This is where Google PageRank came into play. And now the third phase is Trusted Links, which in very simple terms is some links are worth more than others, only after being check to help eliminate potential spamming.

Now we are getting to the really good stuff. Matt Cutts (Google Engineer) made a posting talking about the Google Sandbox, and the Google Algorithm may affect a number of sites under some circumstances, and website owners were seeing this as being Sandboxed (although he says there is no sandbox). An overview of the postings can be found at ThreadWatch. This essentially tells us there is some sort of dampening factor that can affect sites, Google didn’t intentionally do this on purpose, but after reviewing the results, they do like it.

With TrustRank becoming more popular, it seems that the move from the Term Sandbox to TrustBox could be catching on, and from Google Engineer Matt Cutts himself, there is some sort of dampening that can affect some sites.

As of December 6, 2012

Matt Cutts on TrustRank “Matt Cutts: no such thing as Trust Rank https://www.seroundtable.com/google-trust-14416.html Several algorithm factors contribute to what could be classified as “trust””

Other Related Articles on Trust

TrustRank & Recent Google Update

http://www.seroundtable.com/archives/017396.html

What Is Google’s Trust Algorithm in Search?

http://www.seroundtable.com/archives/002010.html