Pairwise metrics
The definition as the field writes it comes first, weaknesses and all. What Entity Bind does about it comes last, in its own paragraph, so you can take one without the other. Filed under measurement.
How you tell whether any of it worked, and which numbers may not be quoted without which other numbers.
every figure, with its population →What pairwise metrics means
Pairwise evaluation enumerates the pairs a system judged, compares each verdict against a label, and computes precision, recall and F1 over those verdicts. It is the natural unit for a matcher, it is cheap to label, and it is what a continuous-integration gate can run on every commit.
Its blind spot is transitivity. Clusters are built from pairwise decisions, and one false edge between two clean groups produces a merged cluster in which a very large number of record pairs are now wrong even though only one decision was. Conversely a system can score well pairwise while fragmenting an entity across three clusters, because the pairs it never compared are not counted against it.
Pairwise metrics are also sensitive to the pair population. Precision measured over the pairs a blocker generated is a different quantity from precision over all pairs, and the two are not comparable across vendors whose blockers differ.
Entity Bind's house rule is that a pairwise figure is never published without the cluster figure beside it. The gate of 2026-09-05 reports pairwise precision 1.0000 (357/357) at 63.10% automation and end-to-end resolution of 70.78% (499/705); End-to-end resolution on the internal gated acquisition fixture with offline stub models. All 499 accepted matches were correct in this fixture. The current 99% product-target diagnostic fails; this run does not measure live model conversion or independently establish calibration. Its gold set is 1,176 labelled company pairs — 216 handwritten adversarial, 960 generated from a fixed seed — plus a 705-record fixture over 250 UK-registered companies.
Where this shows up in Entity Bind
Three pages where this term stops being vocabulary and starts changing an answer, a payload or a price.
- How to measure entity resolution accuracyThere is no single accuracy number for an entity resolution system, and any vendor quoting one is quoting a property of their fixture rather than of their engine.Open
- The accuracy methodEvery published rate with the population it was measured over, its interval and its fixture — including the input shapes that fail.Open
- The audit packThe fixtures, the verdict logs and the replay path: what an evaluator needs to re-run a published rate for themselves.Open
The fastest way past a definition is a file: which of your rows are the same company, which of your existing matches we would not have made, and what we could not decide.
Start freeThe terms next to this one
Each with its own sentence, because the next question is usually about the word beside this one.
Cluster metrics
Cluster metrics evaluate the groups a resolver produced rather than the pairs it judged, which is the only way to see errors that only appear once pairwise decisions are chained into entities.
B-cubed (B³)
B-cubed is a cluster-level evaluation metric that scores precision and recall from the point of view of each individual record, asking what share of its assigned cluster genuinely belongs with it and what share of its true family it was placed with.
Precision
Precision is the share of the matches a system asserted that were correct — and it is meaningless without the share of records the system was willing to decide at all.
Recall
Recall is the share of the true matches present in the data that a system actually found, and it is the metric that abstention costs you.
F1 score
F1 is the harmonic mean of precision and recall, reported as a single number that is only interpretable once you know which of its two components is doing the work.
Cluster
A cluster is the set of records a resolver has decided belong to one entity, and the object that cluster-level metrics are scored against.
The other 13 terms on this shelf
How you tell whether any of it worked, and which numbers may not be quoted without which other numbers.
Internal fixtures only. The historical 2026-08-29 per-tier recording covers 317 GB companies with identifier-rich arrivals seeding the catalogue first; the context tier includes country and registered-town prose. Its legacy replay has no full-prompt-verified calls and cannot certify current calibration. The September 5 CI reports cover 1,176 labelled pairs and 705 records over 250 companies using offline stubs. These are not independent customer holdouts or live production validation; the 99% minimum-input product target remains unmet. Scope is company and organisation identity only — not people, vessels, bank-account ownership, sanctions or emissions. Results may not be used for eligibility decisions about individuals, including employment, credit and tenancy.
Start with your own company records.
Create your workspace, verify your email and run your first company file. The Free plan needs no card. Create API keys in your workspace when you need an integration.
300 credits. No card.
For a guided review of your file, explore the Merge Safety Sprint →