Separate find from check
Use company-name search when the identifier is missing; use number search when the field exists but needs a company match.
VENDOR DATA QUALITY
Vendor master data cleansing for VAT and company records.
Clean vendor master data before an ERP migration. Reconcile legal names, VAT numbers, EINs and duplicate company records with traceable matching evidence.
START WITHExisting vendor tax ID or legal company nameREVIEWMatched entity + fields for reconciliation
Design the control around your systems and policy.Define countries, inputs, match rules, exceptions, evidence and delivery before integration.
Reviewed
THE PRACTICAL ANSWER
How do you cleanse vendor master data?
Profile missing identifiers, inconsistent names and suspected duplicates; resolve the legal company behind each record; then approve field-level corrections with their supporting evidence. VATFind helps research VAT numbers, EINs, registration numbers and company details. Separate automatic formatting changes from entity merges, which need stronger evidence. Retain the original vendor ID and values so an ERP migration or cleanup can be reviewed and reversed.
- An ERP migration exposes missing or inconsistent supplier tax IDs.
- Duplicate vendor records contain different legal names or numbers.
- A data steward needs to research exceptions that automated rules cannot resolve safely.
THE WORKFLOW
Clean the vendor master in four steps
Compare stable identifiers
Use legal name, registration number, country, and address together instead of matching on one field.
Record the source
Keep the returned source and retrieval time with the proposed correction so another reviewer can follow the decision.
Route ambiguous cases
Do not auto-merge records when identifiers conflict. Send them to a person or the appropriate source for review.
WHAT VATFIND DOES
VATFind verifies which legal company is associated with the VAT number or EIN.
VATFind takes the submitted identifier and country, matches it to a source-labelled company record, and returns the legal name, registration number, address, company context, source and retrieval time. The workflow receives an explicit match, discrepancy, ambiguity, no-match or unavailable state instead of a generic pass/fail answer.
Evidence meaning: a verified association means the submitted VAT number or EIN is linked to the returned company record. It is not, by itself, confirmation of current tax-registration status, tax compliance or approval of the wider transaction.
For API checks, use vat_number outside the US and tax_identifier for US EINs. Optional fields can be absent when a source does not provide them.
- 01Available tax ID
- Useful for filling a blank only after the legal company is matched.
- 02Company registration number
- A second identifier for duplicate detection and entity resolution.
- 03Canonical legal name
- The source-reported company name for reconciliation.
- 04Country, region, and address
- Context for separating local branches and similarly named entities.
- 05Source and time
- Evidence for the proposed vendor-master update.
CHOOSE THE RIGHT APPROACH
Cleansing, enrichment and deduplication solve different problems
Cleansing
Correct inconsistent formatting, misplaced identifiers and conflicting values in records you already hold.
Enrichment
Add a missing field from a company record once the business has been resolved, retaining its source and retrieval time.
Deduplication
Decide whether two vendor records represent the same legal entity. A similar name or shared address is not enough to merge them.
CONTROL OWNERSHIP
Give each team one clear responsibility.
A reliable workflow names who owns the input, who interprets the company evidence and who makes the wider decision. That separation prevents a data result from becoming an unexplained approval.
Master-data management
Define canonical fields, survivorship rules and evidence requirements.
ERP programme
Prevent unresolved duplicates and identifiers from migrating into the target system.
Finance operations
Resolve exceptions using business documents and company evidence.
DECISION DESIGN
Define the evidence and action at each stage.
| Stage | Evidence to retain | Controlled action |
|---|---|---|
| Profile | Completeness, duplicate and conflict rates by country and identifier type. | Create an exception taxonomy before changing records. |
| Resolve | Legal name, registration number, tax ID, address, source and match state. | Propose corrections with confidence and evidence rather than overwriting values. |
| Publish | Approved canonical value, original value, reviewer and effective time. | Update downstream systems and retain a reversible change history. |
EXAMPLE WORKFLOW
One supplier appears three times before an ERP migration
A migration file contains a legal company, a shortened name and a group brand as separate vendors. Compare registration country and company number first, then use the available tax IDs and addresses to investigate the suspected duplicates. Two records may belong to the same legal entity while the third is a subsidiary that must remain separate. Prepare a proposed merge with the original vendor IDs, evidence and affected transactions for the data steward's approval.
See the API requests and responsesIMPLEMENTATION CHECKLIST
Vendor master data cleansing checklist
- Separate null, malformed, conflicting and unverified values into different queues.
- Use country-specific identifier rules instead of one global regular expression.
- Create deterministic joins on typed identifiers before using fuzzy-name matching.
- Require human review for cross-entity merges and identifier conflicts.
- Measure data quality again after the migration or remediation load.
WHAT TO MEASURE
Prove the control is useful.
Track operating quality rather than claiming that a company match guarantees the downstream outcome.
- Duplicate vendor records removed with evidence
- Identifier completeness by country
- Manual-review volume and false-merge rate
READ THE RESULT CORRECTLY
The public lookup is an exception-review tool, not a bulk processor.
VATFind supports individual searches and saved lookup history. It does not currently accept a CSV file or promise automated bulk validation. For volume workflows, request API access and define the source, matching, exception, and audit requirements first.
Review data and limitationsFREQUENTLY ASKED QUESTIONS
Vendor data cleansing questions
Should legal names be overwritten automatically?
No when the new name conflicts with documents or other identifiers. Preserve the original value, record the proposed source, and send ambiguous cases for review.
Can I use VAT number alone as a duplicate key?
It can be a strong signal, but not every record supplies one and identifiers can be entered incorrectly. Combine it with registration number, legal name, country, and address.
Does VATFind support a bulk CSV upload?
Not in the current public product. The page describes the exception-review use case; it does not claim a batch feature that is not available.
What should an API data-cleanup project define?
Agree the supported countries and identifiers, matching thresholds, source priority, exception rules, evidence fields, refresh cadence, and handling of no-match results before implementation.
Is vendor data cleansing the same as vendor master data management?
Cleansing repairs a defined set of data-quality problems. Vendor master data management adds the ongoing ownership, update rules, system mapping and monitoring needed to keep the records usable after that cleanup.
How should a company name with no VAT number be processed?
Use the name-search route with the registration country. Compare the company details before proposing an available tax ID. Keep unresolved cases separate from records whose identifiers have been confirmed, and never substitute another company's number.