Account enrichment data intake without invented certainty
Review account enrichment data for provenance, confidence, field authority, and safe activation before it changes GTM workflows.
Account enrichment data intake
Treat enrichment as a dated external observation, not automatic truth. Before import, define which fields may be added, suggested, or overwritten and who owns each decision.
Record vendor, dataset, collection or refresh date, match method, confidence, licensed use, covered population, and source field. Preserve raw values separately from normalized CRM values.
Define the permitted field map
List destination fields, formats, allowed values, authority, fallback, and overwrite behavior. Separate firmographic attributes from contact details, intent signals, technology observations, and derived classifications.
A convenient destination field may carry a different definition. Stop the mapping when units, time periods, company hierarchy, or category meaning cannot be reconciled honestly.
Test identity matching
Sample exact matches, domain matches, subsidiaries, parent companies, renamed organizations, shared domains, and ambiguous records. Measure outcomes against reviewed examples without turning uncertain matches into confirmed accounts.
Keep unmatched and disputed rows in a controlled queue. An enrichment vendor identifier should remain available so a correction can be traced back to the delivered record.
Resolve conflicts by field authority
When enrichment differs from customer, seller, billing, support, or product evidence, use the documented authority for that field and purpose. Preserve the conflict and observation dates.
Do not overwrite verified first-party information merely because the external value is newer. Newness and authority answer different questions, especially for ownership, company size, region, and lifecycle data.
Activate changes in stages
Load a test population, verify transformations, and inspect routing, scoring, segmentation, personalization, reporting, and suppression effects. Expand only after owners accept the observed behavior.
Keep rollback data and the exact import version. If a change creates unexpected outreach or ownership, pause activation and correct the affected records before processing the remaining set.
Monitor quality after intake
Review corrections, unmapped values, match failures, overrides, stale observations, and workflow exceptions by field and source. Send evidence-backed defects to the vendor or internal data owner.
Renew or change the feed based on fitness for defined decisions, not on fill rate alone. Complete fields can still be misleading when provenance or confidence is hidden.
Sources
- Data Quality, Salesforce Trailhead. Covers data quality planning, assessment, and governance concepts for CRM records.
Related GTM guides
Frequently asked questions
Should enrichment overwrite CRM data automatically?
Only where an approved field policy permits it and testing shows the downstream effects. Many fields should be suggested or reviewed first.
What should happen to uncertain account matches?
Retain the source record and confidence, then route the match to an owned exception path without activating it as confirmed.