About Link Signals
Link Signals is an independent site about backlink data — the metrics, the tools that produce them, and what you’re entitled to conclude from them.
It is not a link-building blog, not a tool review site, and not an agency. There’s nothing here to buy and no affiliate links.
Why measurement specifically
Because the interpretation layer is where most link analysis goes wrong, and almost nobody writes about it.
The tools are genuinely good. Ahrefs, Semrush, Moz and Search Console each expose enormous amounts of data about who links to whom. What they can’t do is tell you what their numbers don’t cover — the pages their crawler never reached, the links that died three months ago and haven’t been recrawled, the deduplication rule that turns eight hundred sitewide footer links into one row or eight hundred depending on the vendor.
So a reasonable person opens two tools, sees 1,400 backlinks in one and 5,900 in the other, and concludes that one of them is broken. Neither is. They’re reporting different crawls under different rules, and knowing that changes what you do next.
What we think
A metric is a procedure, not a property. Domain Rating isn’t a thing your site has. It’s the output of a computation over a particular index at a particular moment, on a scale that is almost certainly logarithmic. Every sentence you write about it should survive that substitution.
Most link metrics are ordinal at best. They’re useful for ranking things against each other — this prospect is stronger than that one — and much weaker as absolute quantities. “We grew DR from 40 to 45” is a sentence whose meaning depends entirely on curve shape you can’t see.
Attribution is genuinely hard and pretending otherwise is the real error. A link landed, rankings moved, and somewhere in the same fortnight so did your content, your competitors, your internal linking, and the algorithm. You can build an honest case. You cannot isolate one variable in a live search index, and reports that claim to have done so are usually the ones counting the link twice.
Correlation studies deserve more suspicion than they get. Correlating a vendor’s own metric with rankings tends to produce a satisfying number and very little information, partly because the metric was built from link data in the first place.
On sources
We don’t publish statistics we can’t point at. No invented correlations, no “studies show” without a named, public study, no round numbers manufactured to make a paragraph land.
Worked examples appear throughout — they’re the clearest way to explain a scale or a discrepancy — and they are always labelled as hypothetical. A site arguing for metric honesty that fabricates its own figures would not be arguing for much.
We also don’t state the internals of anyone’s proprietary formula as fact. Where something is documented by the vendor, that’s said. Where it’s observable, that’s said. Where it’s inference, that’s said too.
Elsewhere in this network
Link acquisition — earning links, digital PR, audits, disavow — is covered at backlink.so. The technical side — redirects, canonicals, internal linking, migrations — is at backlink.sh. This site stays on the numbers.