What columns does the file need?

At minimum a legal name and an entity type. Country, date of birth for individuals and a registration number for companies raise match quality sharply, because they let a reviewer dismiss common-name candidates in seconds instead of minutes.

Should people and companies go in the same file?

Separate them where the tool allows it. Person and entity matching lean on different signals, dates of birth and nationality against registration numbers and addresses, so mixing them produces noisier candidates for no gain.

What do we keep after the run?

The input file, the results export, the list versions and date used, the candidate detail per row, and the analyst disposition with a reason. That set is what makes the run reproducible months later, which is the only reason to keep any of it.

SCREENING GLOSSARY · GUIDEUPDATED 2026-09-04

Screening Glossary

What is batch screening?

Compliance duties apply to a portfolio, not only to the next new customer. Batch screening checks many names at once, a customer book, a supplier file, a payroll or an investor register, against sanctions, PEP and watchlist data in a single run.

Batch runs are used for periodic re-screening, for onboarding a book after an acquisition, and for the first pass over a supplier base nobody ever checked. Almost all of the difficulty is in the input data: missing country codes, persons and companies mixed in one column, trading names where the legal name was needed. Poor input inflates false positives and hides genuine matches, and no amount of matching sophistication compensates for it.

Run screening View pricing

What this workflow covers

SCOPE

Give every row a legal name, an entity type and a country before you run anything.
Triage candidates by confidence, then export row-level results with a disposition against each row.
Deduplicate first, or repeated rows for the same party will distort your false-positive statistics.
Batch runs here accept a CSV or XLSX upload or an API call, bill at the same per-check rate as a single search, and return a row-level export.
A batch run is not a weaker check. The matching is the same; the volume and the review workflow are what differ.

Compliance glossary

TERMS

Row-level evidence: A per-record output showing what was screened, which lists were checked, which candidates were returned and how each was dispositioned.
Deduplication: Collapsing repeated records for the same party before screening, so candidate counts and false-positive rates stay meaningful.

Authoritative references

SOURCES

01
A Framework for OFAC Compliance Commitments
U.S. Department of the Treasury — OFAC
02
The FATF Recommendations — Recommendation 10 (Ongoing due diligence)
Financial Action Task Force

Frequently asked questions

Q&A

What columns does the file need?: At minimum a legal name and an entity type. Country, date of birth for individuals and a registration number for companies raise match quality sharply, because they let a reviewer dismiss common-name candidates in seconds instead of minutes.
Should people and companies go in the same file?: Separate them where the tool allows it. Person and entity matching lean on different signals, dates of birth and nationality against registration numbers and addresses, so mixing them produces noisier candidates for no gain.
What do we keep after the run?: The input file, the results export, the list versions and date used, the candidate detail per row, and the analyst disposition with a reason. That set is what makes the run reproducible months later, which is the only reason to keep any of it.

Screening Glossary Ongoing monitoring False positive Audit trail Batch screening Customer screening Sanctions screening