Trust is commonly understood to be a standalone quality that is passed between sites regardless of link neighborhood or topical vertical. What I’m going to demonstrate is that “trust” is not a thing that trickles down from a trusted site to another site. The implication for link building is that many may have been focusing on the wrong thing.
Six years ago I was the first person to write about link distance ranking algorithms that are a way to create a map of the Internet that begins with a group of sites that are judged to be trustworthy. These sites are called the seed set. The seed set links to other sites, which in turn link to ever increasing groups of other sites. The sites closer to the original seed set tend to be trustworthy websites. The sites that are furthest away from the seed set tend to be not trustworthy.
Google still counts links as part of the ranking process so it’s likely that there continues to be a seed set that is considered trustworthy from which the further away you a site is linked from the seeds the likelier it is considered to be spam.
Circling back to the idea of trust as a ranking related factor, trust is not a thing that is passed from one site to another. Trust, in this context, is not even a part of the conversation. Sites are said to be trustworthy by the link distance between the site in question and the original seed set. So you see, there is no trust that is conveyed from one site or another.
The word Trustworthiness is even a part of the E-E-A-T standard of what constitutes a quality website. So trust should never be considered as a thing that is passed from one site to another because it does not exist.
The takeaway is that link building decisions based on the idea of trust propagated through links are built on an outdated premise. What matters is whether a site sits close to trusted seed sites within the same topical neighborhood, not whether it receives a link from a widely recognized or authoritative domain. This insight transforms link evaluation into a relevance problem rather than a reputation problem. This insight should encourage site owners to focus on earning links that reinforce topical alignment instead of chasing links that appear impressive but have little, if any, ranking value.
Why Third Party Authority Metrics Are Inaccurate
The second thing about the link distance ranking algorithms that I think is quite cool and elegant is that websites naturally coalesce around each other according to their topics. Some topics are highly linked and some, like various business association verticals, are not well linked at all. The consequence is that those poorly linked sites that are nevertheless close to the original seed set do not acquire much “link equity” because their link neighborhoods are so small.
What that means is that a low-linked vertical can be a part of the original seed set and display low third-party authority metrics scores. The implication is that the third-party link metrics that measure how many inbound links a site has fail. They fail because third-party authority metrics follow the old and outdated PageRank scoring method that counts the amount of inbound links a site has. PageRank was created around 1998 and is so old that the patent on it has expired.
The seed set paradigm does not measure inbound links. It measures the distance from sites that are judged to be trustworthy. That has nothing to do with how many links those seed set sites have and everything to do with them being trustworthy, which is a subjective judgment.
That’s why I say that third-party link authority metrics are outdated. They don’t follow the seed set paradigm, they follow the old and outdated PageRank paradigm.
The insight to take away from this is that many highly trustworthy sites are being overlooked for link building purposes because link builders are judging the quality of a site by outdated metrics that incorrectly devalue sites in verticals that aren’t well linked but are actually very close to the trustworthy seed set.
The Important Of Link Neighborhoods
Let’s circle back to the observation that websites tend to naturally link to other sites that are on the same topic. What’s interesting about this is that the seed sets can be chosen according to topic verticals. Some verticals have a lot of inbound links and some verticals are in their own little corner of the Internet and aren’t link to from outside of their clique.
A link distance ranking algorithm can thus be used to calculate the relevance according to whatever neighbhorhood a site is located in. Majestic does something like that with their Trust Flow and Topical Trust Flow metrics that actually start with trusted seed sites. Topical Trust Flow breaks that score down into specific topic categories. The Topical Trust Flow metric shows how relevant a website is for a given metric.
My point isn’t that you should use that metric, although I think it’s the best one available today. The point is that there is no context for thinking about trustworthiness as something that spreads from link to link.
Once you can think of links in the paradigm of distance within a topic category it becomes easier to understand why a link from a university website or some other so-called “high trust” site isn’t necessarily that good or useful. I know for certain because there was a time before distance ranking where the topic of the site didn’t matter but now it does matter very much and it has mattered for a long time now.
The takeaways here are:
- It is counterproductive to go after so-called “high trust” links from verticals that are well outside of the topic of the website you’re trying to get a link to.
- This means that it’s more important to get links from sites that are in the right topic or from a context that exactly matches the topic, from a website that’s in an adjacent topical category.
For example, a site like The Washington Post is not a part of the Credit Repair niche. Any “trust” that may be calculated from a New York Times link to a Credit Repair site will likely be dampened to zero. Of course it will. Remember, seed set trust distance is calculated within groups within a niche. There is no trust passed from one link to another link. It is only the distance that is counted.
Logically, it makes sense to assume that there will be no validating effect between irrelevant sites. relevant website for the purposes of the seed set trust calculations.
Takeaways
- Trust is not something that’s passed by links
Link distance ranking algorithms do not deal with “trust.” They only measure how close a site is to a trusted seed set within a topic. - Link distance matters more than link volume
Ranking systems based on link distance assess proximity to trusted seed sites, not how many inbound links a site has. - Topic-based link neighborhoods shape relevance
Websites naturally cluster by topic, and link value is likely evaluated within those topical clusters rather than across the entire web. A non-relevant link can still have some small value but irrelevant links stopped working almost twenty years ago. - Third-party authority metrics are misaligned with modern link ranking systems
Some third-party metrics rely on outdated Page Rank-style link counting and fail to account for seed set distance and topical context. - Low-link verticals are undervalued by SEOs
Entire niches that are lightly linked can still sit close to trusted seed sets, yet appear weak in third-party metrics, causing them to be overlooked in link builders. - Relevance outweighs perceived link strength
Links from well-known but topically irrelevant sites likely contribute little or nothing compared to links from closely related or adjacent topic sites.
Modern link evaluation is about topical proximity, not “trust” or raw link counts. Search systems measure how close a site is to trusted seed sites within its own topic neighborhood, which means relevant links from smaller, niche sites can matter more than links from famous but unrelated domains.
This knowledge should enable smarter link building by focusing efforts on contextually relevant websites that may actually strengthen relevance and rankings, instead of chasing outdated link authority scores that no longer reflect how search works.
Featured Image by Shutterstock/Kues
Simply SUBSCRIBE to read this article (it's free)
