RESEARCH

Measurements we made ourselves, published with the working

There is a lot of received wisdom in link building and very little of it carries a sample size. The studies here exist because the scanner makes a specific kind of measurement cheap: opening a large number of exact pages and recording what is on them. Where that produces something worth knowing, it gets written up in full, including the parts that do not flatter the product.

PUBLISHED

Studies

Published 7 September 2026 · 1,001 rows · Verified by hand

We opened 1,001 purchased backlink prospects. Domain Rating sorted them backwards.

964 of the 1,001 rows could not carry a followed comment link, and 818 had no comment form at all. The finding worth your attention is not that number but the sort order: not one of the 285 rows at Ahrefs DR 50 or above was among the 37 usable ones, because the high scores were hosted subdomains inheriting a rating that belonged to their host.

Rows checked: 1,001
Pages readable: 788 · 78.7%
Comment form open: 183 · 18.3%
Would carry a followed link: 37 · 3.7%

Read the study, the method and the limits

One study is one study. The next one will be run when there is a question worth the scan time, and it will be listed here whether or not the answer is convenient.

HOW WE PUBLISH

Four rules, and the reason each one exists

Every number arrives with its denominator

Not "96.3% of prospects", but "964 of 1,001 rows on one purchased list". A percentage separated from its sample can be quoted into meaning almost anything, which is exactly why it gets quoted.

Every number arrives with its date

A comment form is open on a date. Pages change. A finding without a date is a claim about the present that nobody re-checked.

Every number arrives with its limit

One list is one list. Where a result cannot generalise, the page says so in the same breath as the result, not in a footnote under the chart.

The raw rows are published

The tab-separated data and the script that aggregates it are in the repository. If the arithmetic here is wrong, it is checkable rather than arguable.

Quoting any of this

Use it. The one thing we ask is that a figure keeps the denominator and the date it was measured on, because that is what makes it checkable rather than merely repeatable. If you find an error, write to us and the correction goes on the page with its own date.

SUGGESTED CITATION

Mehdi, H. (7 September 2026). We opened 1,001 purchased backlink prospects. WebDiagnosis.