What is TrustRank?

TrustRank is a link analysis method that separates trustworthy web pages from spam by starting with a small set of pages that human experts have checked and passing trust outward through their links. It was published in 2004 by researchers at Stanford University and Yahoo.

TrustRank is a method for separating useful web pages from spam. Experts first review a small set of "seed" pages and mark the trustworthy ones. The algorithm then follows links out from those seeds, on the idea that good pages rarely link to spam.

Where does TrustRank come from?

TrustRank was introduced in the 2004 paper "Combating Web Spam with TrustRank" by Zoltán Gyöngyi and Hector Garcia-Molina of Stanford University and Jan Pedersen of Yahoo. They presented it at the VLDB conference that year. It was a research answer to link spam and spamdexing, which were inflating search results at the time.

How does TrustRank work?

  1. Pick seed pages. A small set of pages is chosen and reviewed by a human expert.
  2. Mark the good ones. Pages the expert judges reputable become the trusted seeds.
  3. Spread trust through links. Pages that the seeds link to receive some trust, and pages those pages link to receive a little less.
  4. Let trust fade with distance. The further a page sits from the seed set, the less reliable its trust score becomes.

Because it starts with human judgement and then runs automatically, TrustRank is a semi-automatic method. A related idea, sometimes called Anti-Trust Rank, works in reverse: pages close to known spam are more likely to be spam.

Does Google use TrustRank?

Google has not confirmed using the TrustRank algorithm. Two things often get mixed up:

  • The research paper. TrustRank came from Stanford and Yahoo, not Google.
  • Google's patent. Google holds a separate patent, "Search result ranking based on trust", filed in 2006 and granted in 2009. It describes adjusting result scores using the trust of the people who label documents. A patent shows an idea Google protected, not what runs in Search today.

When asked about a trust metric in a 2021 office-hours session, Google's John Mueller said: "I don't think we have like a trust factor that we can look at."

What Google does document is trust as a quality concept. Its guidance on helpful content explains E-E-A-T (experience, expertise, authoritativeness and trustworthiness) and says "trust is most important" of the four. The same page states that E-E-A-T itself is not a specific ranking factor.

Can you check your TrustRank?

No. Google does not publish a trust score for any website. SEO tools that show a trust or authority number build their own estimate from their own crawl of backlinks. Metrics like domain authority are useful for comparing sites inside one tool, but they are not Google data.

What this means for B2B marketers

You can't optimise for a hidden score, but you can do what the method rewards:

  • Earn links from reputable, relevant sites in your industry.
  • Avoid link schemes and links from spammy neighbourhoods.
  • Show who wrote your content and why they know the subject.
  • Cite sources for claims, especially statistics and product comparisons.

Frequently asked questions

Does Google use TrustRank?

Google has not confirmed using the TrustRank algorithm from the 2004 paper. It does hold a separate 2009 patent called "Search result ranking based on trust", but a patent does not show what runs in Search today. Google's John Mueller said in 2021 that he did not think Google has a trust factor it can look at.

Can I check my site's TrustRank?

No. Google does not publish a trust score for any site. Tools that show a "trust" or "authority" number calculate their own estimate from their own link data, so treat it as a rough comparison, not a Google metric.

How do I make my site more trustworthy to search engines?

Google's guidance on helpful content says trust is the most important part of E-E-A-T. Accurate content, clear authorship, honest sourcing and links earned without manipulation all support it.