Field guides / methodology
METHODOLOGY · COMPANY RELATIONSHIPS

How to evaluate corporate group data.

Check the legal identity, relationship evidence and coverage limits before using a corporate hierarchy in an enterprise account plan.

Evaluate your customer groups
01

Resolve the legal identity first.

Use a legal name, country and registration identifier where available. A website or brand can represent several group companies. Review candidates when the match is ambiguous, and retain the selected entity’s identifier with your account plan.

02

Keep observed relationships and interpretation distinct.

Preserve source classifications and dates when supplied. A relationship can be registry-backed or modelled; a corporate connection is not an assertion that a company needs your product. Missing classification should remain unknown rather than being relabelled as verified.

03

Count records, legal companies and locations separately.

Group results can contain head offices, single-location records and branches. Count unique returned identifiers separately from row counts. Confirm the record type before describing a record as a separate subsidiary or addressable customer.

04

Measure coverage on a representative sample.

Choose familiar customer groups across your markets, sectors and entity sizes. Compare expected relationships with returned companies, missing countries, unresolved profiles and reported fields. A worldwide country list is not a guarantee that every company or field is available.

  • Record your expected entities independently before running the test.
  • Use known legal identifiers and preserve the original response.
  • Report missing and unresolved records alongside matches.
  • Record the retrieval date and any result limits.
05

Do not infer completeness from the top of a tree.

The highest returned parent may be the top of the available result rather than the verified global ultimate parent. Missing subsidiaries can reflect limited disclosure, unresolved links, source availability or changes not yet returned. Absence is not proof of independence.

06

Separate monitoring signals from business conclusions.

A returned ownership or status change is a reason to review the account. A newly returned company is not automatically newly incorporated or acquired. Check timestamps, evidence and monitoring coverage before changing your plan.

07

Agree data use and delivery scope.

Evaluate the online workspace, permitted exports, REST API and MCP with your intended workflow. Confirm countries, fields, monthly allowances, contact credits, monitoring scope and contractual data-use rights before an enterprise rollout.

THE DETAILS THAT MATTER

Before you put it to work.

How can I assess EntityReach coverage?

Use familiar groups and known registration identifiers, compare expected entities with returned records and record missing fields and links. Country-level coverage is context; the sample test establishes fit for your portfolio.

Why might a subsidiary be missing?

Disclosure, relationship availability, unresolved legal identity, freshness or a bounded result can create a gap. Review the source and result limits rather than treating a missing company as proof that no subsidiary exists.

Does every relationship have registry evidence?

No universal evidence guarantee applies. Preserve the classifications and source information actually returned, and flag unavailable evidence for review.

PUT THE GUIDE TO WORK

Evaluate your customer groups

Evaluate your customer groups