What link velocity measures
Link velocity is the rate at which new links, or more usefully new referring domains, appear over time. Tools plot it as a line: how many domains started linking to you this month compared with last. On its own that line is a description, not a verdict.
The reason anyone watches it is shape. A site that publishes regularly, runs occasional campaigns and gets covered when it does something notable produces an uneven but explainable line — quiet stretches, a rise around a launch or a piece of research, then a settling period as the coverage is quoted and syndicated. A site buying links produces something different: an even drip that begins on an invoice date and stops when the contract does.
Why link velocity matters
It matters as a diagnostic, mostly for the person paying. If your supplier’s reporting shows a near-identical number of new domains every month regardless of what you published, they are not earning links, they are placing them. If a competitor’s line jumps sharply, it is worth finding out why, because usually something real happened: a launch, a piece of data, a story that got picked up.
It also helps set expectations. Earned links follow the work, so a month with nothing to say produces few links, and that is normal rather than a failure. Judging a campaign by whether the line rose every single month pushes people towards buying.
Where link velocity goes wrong
The main error is treating it as a lever. There is no published safe rate, no threshold Google confirms, and no reason to think that pacing purchased links makes them acceptable. A slow scheme is still a link scheme; the pace changes nothing about the intent behind it.
The second error is trusting raw link counts. Backlink tools record every URL they find, so one mention republished by scrapers or carried across a newswire can add a great many links from a handful of sources. That looks like a spike and is really a single event. Watching referring domains instead of links removes most of the noise.
How to act on it
Read the line next to a calendar. Mark the months where you published research, launched something, sponsored an event or appeared in the press, then see whether the links line up. Where they do, you have learned what produces coverage and can do more of it. Where a rise has no cause you recognise, find out what it was before celebrating it.
Then stop managing to the metric. Plan the work — the research, the tool, the local partnership, the outreach — and let velocity be the record of it. That is what a link building programme should produce as a by-product rather than as its objective.