Track your inbound link profile, and the sources you are missing
The links pointing at your pages are won and lost without anyone telling you. We read your profile totals every day, keep for each link what its source says about it, and list the domains that cite your competitors without citing you yet.
Access opens in waves: we email you when yours is ready.
Our numbers, with their provenance
-
43signals recorded per link profile computed
-
11signals per inbound link measured
A link profile is judged on its movement
A total of inbound links says almost nothing on its own: what matters is what it gains, what it loses, and where that comes from. The three situations below almost always come before the sentence “our link building has stalled”, said without anyone being able to name the month.
-
01
You hear about lost links months later
An article taken down, a site redesign, a forgotten redirect: the source vanishes quietly, and traffic drops before anyone connects the two events.
-
02
The total climbs, and the cause stays hidden
Three links won on a directory carry different weight from one link won on a publication in your field. Without the breakdown by domain and by anchor text, a rising total can describe a decline.
-
03
Your competitors have sources you have yet to earn
They are public, and finding them by hand means reading three competitor profiles line by line. Nobody does that twice in a row.
What we measure
Four readings, each with its date
Each one carries the target exactly as it was measured: change a project’s address and the series starts fresh, rather than blending two domains into one curve. The monthly history and the daily reading stay on two separate scales.
- The profile, read every day (inbound links, referring domains, pages receiving links, broken links and risk score: the totals go out once a day and form the series you compare. The monthly history is pulled once, on the first pass.)
- Every link, with its signals (the anchor text, the follow status granted or withheld from engines, the rank of the source page and source domain, first and last time it was seen. The number of signals kept per link is published in our claims registry.)
- Referring domains, merged properly (one site reached through two hosts (with and without www) counts as one domain: counters are added up, the highest risk score is kept, and the first appearance keeps its earliest date.)
- The sources your competitors have (on request, the module lists the domains and pages that cite at least one declared competitor of your project without citing you yet. Those competitors come from rank tracking, so from your own readings.)
Further reading
How to analyse an inbound link profile: the method, not the score
The guide: reading an inbound link profile, and deciding what deserves action.
Read ComparisonNessFlow vs Semrush
What an all-in-one suite shows of a link profile, and what its model asks in return.
Read GlossaryGlossary
Inbound link, referring domain, risk score: the definitions.
ReadOther modules
Frequently asked questions
How does link tracking start on a project?
An admin on your team switches it on, project by project. Until that happens the module stays pending: it reads nothing and consumes nothing. This is a spending decision, so it belongs to someone entitled to commit it. Switching it on applies to the project’s domain, and changing that address starts a fresh series rather than blending two profiles into one curve.
How often is the profile read?
Totals go out once a day, detailed lists once a week. A link profile moves slowly enough that a weekly rhythm is enough for the detail, while daily totals are enough to see an inflection. The date of every pass sits next to the number, so you can see whether the curve you are reading is complete.
What happens when a reading fails?
The cell stays empty and the screen shows “not measured”. This distinction carries the whole module: a lost link is a dated fact, while a collection that failed is nothing at all. Writing a zero in its place would sink your curve on the day of a network outage. The gained and lost counters also come from a call separate from the totals: they can stay empty while the rest of the row is properly measured, and the screen says so row by row.
Where do the competitive comparison domains come from?
From the competitors declared for your project in rank tracking, so from domains your own readings surfaced. The module keeps those citing at least one of these competitors without citing you yet, and the pass runs on request, never on its own. What each source is worth to you stays your judgement: the list says who cites whom, on a given date, and the rest belongs to your strategy.
The scope of this tracking, and its limits
Every reading is dated, and tracking is switched on project by project, by an admin on your team. The scope below follows from those two facts, and writing it here beats letting anyone imagine unlimited watching.
What we cover
- Profile totals read every day, with the date of every pass
- The list of links, referring domains and anchor texts, refreshed every week
- Monthly history pulled on the first pass, kept on a scale separate from the daily one
Stated limits
- Switched on by an admin, project by project: a pending project collects nothing and costs nothing
- A missed reading shows as “not measured” and stays distinct from a lost link
- No score out of a hundred: this module carries a trend, which the scorecard leaves out of its calculation
- No disavow file and no outreach: the module measures, earning links stays your work
Your link profile, with the date of every reading
Daily totals, the breakdown by link, domain and anchor text, and the sources your competitors already have: enough to decide where to put the effort.
Access opens in waves: we email you when yours is ready.