ⓘ Reading Time: 5 mins
What are match scores?
Our transaction enrichment process involves searching through our vast, proprietary data set to identify parties who may have participated in the transaction ("counterparties"). When we identify a potential counterparty, we assign them a "match score" – an assessment of how confident we are that the identification is correct. Match scores range in value from 0.0 to 100.0 – the higher the match score, the higher the likelihood that a counterparty was involved in a transaction.
How are match scores generated?
We use a machine learning model to predict match scores. We're consistently working to improve this model and create more accurate predictions.
How should match scores be used?
Match scores can help you make decisions about how to use enriched data in your systems. For example, for decision-making processes (e.g. card authorization flows or fraud assessments) we suggest only using matches with scores of >70, whereas for customer analysis, budgeting, or UX/UI improvements, matches with scores >55 are typically sufficient.
What does a given match score mean?
| Score Range | Interpretation |
|---|---|
| 90‑100 | A very high confidence match – we strongly believe this counterparty was involved in the transaction. |
| 80‑89 | A high confidence match – we believe this counterparty participated in the transaction. |
| 70‑79 | A medium confidence match – some details may differ between the counterparty and transaction but we think this is likely the involved counterparty. |
| 50‑69 | A low confidence match – a relevant counterparty was found in our records, but we aren't certain that it aligns well with the transaction information. |
| 0‑50 | A very low confidence match – this may be the counterparty in the transaction, but the information doesn't match our records well. |
null | No counterparty match was found in our records. |
*Note that as our model improves and becomes more accurate, these guidelines may be adjusted.
What if there is no match score?
In the rare event we are unable to match a transaction to an entity in our backing data we will return a null match score indicating that the returned data is not based on a match and is instead derived from what was in the request.