The distribution
Trust Flow across all 22,260 referring domains ever recorded:
| Trust Flow | Domains | Share |
|---|---|---|
| 0 | 15,421 | 69.3% |
| 1–10 | 4,031 | 18.1% |
| 11–20 | 1,318 | 5.9% |
| 21–30 | 647 | 2.9% |
| 31–40 | 397 | 1.8% |
| 41–50 | 237 | 1.1% |
| 51–60 | 86 | 0.4% |
| 61–70 | 53 | 0.2% |
| 71–80 | 36 | 0.2% |
| 81–90 | 23 | 0.1% |
| 91–100 | 11 | 0.05% |
Seven in ten referring domains are Trust Flow 0. Two percent are Trust Flow 41 or above. On a profile where not one link was ever bought.
Every one of those 15,421 zero-trust domains arrived unsolicited — scrapers, aggregators, expired-domain churn, auto-generated directories, comment spam, sites that copied a post wholesale. That is the natural background radiation of having published on the internet for twenty years.
It defuses disavow panic
A large proportion of the alarm generated by backlink audits rests on the discovery of exactly this kind of domain. A report showing thousands of zero-authority referring domains reads as evidence of a problem: a negative SEO campaign, a legacy of poor vendor choices, something requiring urgent remediation and a disavow file.
It is worth having a calibration point before accepting that framing. Here is a profile in which nothing was ever bought, and 69.3% of the referring domains are Trust Flow 0. Junk links are what the internet does to anyone who publishes on it. They are not, by themselves, evidence of anything at all.
None of which means disavow is never appropriate. It means the trigger for considering it is a manual action or a specific, identifiable pattern — not the raw presence of low-authority domains, which every site of any age has in quantity.
It calibrates what a proposal can promise
The same table works in the other direction.
In twenty years, this domain earned 446 referring domains at Trust Flow 41 or above — approximately 22 per year, on a site with genuine industry standing, publishing continuously, in a niche where the author is known.
A proposal promising fifty links at that quality within a quarter is describing an outcome roughly double what this site achieved annually, compressed into a tenth of the time. That is not an ambitious target. It is a description of something manufactured, and the arithmetic is worth doing before the contract rather than after.
What natural anchor text looks like
The top anchor across the profile is the brand: “bill hartzer” at 272,856 links, about 21% of the profile on the name alone. Below that sit naked URLs, the site name, and post titles.
There is no commercial exact-match anchor anywhere near the top. Not one.
This is worth knowing because anchor text ratios are routinely presented as something to engineer toward a target distribution. What an unengineered distribution actually looks like is: overwhelmingly brand, then URLs, then whatever the linking author happened to call the page. If your profile needs a spreadsheet to manage its anchor ratios, that is itself the signal.
Telling a manufactured profile from a natural one
If junk links are normal, what actually distinguishes a bought profile? Not the presence of low-authority domains, which this page has just established is universal. Four things, in rough order of reliability:
Metric signatures that repeat. Natural referring domains have scattered, uncorrelated metrics. A cluster of nominally independent sites sharing near identical Trust Flow and Citation Flow values — the same numbers to the point, across several domains with unrelated topics — is one operation wearing several hats. This is the single most reliable tell, and it is why reading the graph beats reading a list.
Anchor distribution that no human would produce. Natural anchors are dominated by brand, naked URLs and page titles, as they are here. A profile whose top anchors are commercial exact-match phrases is describing what somebody wanted to rank for, not what anybody chose to call the page.
Acquisition patterns with no editorial rhythm. Links earned by publishing arrive when something gets noticed — irregularly, in bursts tied to specific pieces. Links acquired at a constant rate, or in a single step change, are describing a purchase schedule.
Infrastructure overlap. Shared subnets, shared registrant details, shared analytics identifiers and identical CMS fingerprints across supposedly separate publishers. Slower to check and hard to argue with once found.
Notice that none of these is a metric threshold, and none of them can be evaluated from a sorted list of referring domains. That is the actual lesson of this page: the signal is in relationships between domains, not in the properties of any one of them.
Fresh versus Historic, and why the gap is not decay
The Fresh export contains 11,815 links from 1,130 domains — 0.9% of the links and 5.1% of the domains in the Historic file.
That gap gets read as “99% of my links are gone.” It is not that. Historic is a cumulative record of every link ever seen since 2006; Fresh is what has been crawled recently. A link can be perfectly alive and absent from Fresh simply because its page has not been recrawled lately.
The honest decay number is the one on the survival page: 49.6% of all links ever recorded are flagged lost. The Fresh-to-Historic ratio is a statement about crawl windows.
Worth naming as its own point, because the conflation is common and it is precisely how an audit manufactures alarm: quote the Historic total, then the Fresh total, then let the reader draw the conclusion.
Questions
Should I disavow low Trust Flow links?
Not on the basis of Trust Flow alone. This profile has 15,421 Trust Flow 0 referring domains and nothing was ever bought. The case for a disavow file is a manual action, or an identifiable pattern you can describe, not a metric threshold.
How did 15,000 junk domains find one site?
Scrapers republish content with links intact. Aggregators pull feeds. Expired domains get rebuilt with the old outbound links still in place. Auto-generated directories harvest from other directories. None of it requires anyone to have done anything.
Is 22 quality domains a year good?
It is what continuous publishing in a niche with genuine standing produced without any link acquisition effort at all. Deliberate work should beat it. The point is the order of magnitude: dozens per year is a real outcome, hundreds per quarter is a claim that needs explaining.