Study first
Review the ideas behind the questions
These notes connect cleanup work to the governance choices that stop bad CRM data from returning.
Define identity before cleanup
Good CRM cleanup starts with knowing which fields reliably identify a record and which records should stay distinct.
- Use stable unique identifiers when importing or updating records so new rows do not create avoidable duplicates.
- Review duplicate candidates before merging because ownership, associations, and activity history can affect which record should remain.
- Document the source system and matching rule behind cleanup work so the same problem does not return in the next import.
Prevent bad data at entry
Data quality is easier to protect when fields, validation, and process rules match how teams actually work.
- Required fields should support a real process decision, not just patch one dashboard after the fact.
- Validation rules are strongest when they block clearly invalid values and give users enough direction to fix the record.
- Governed picklists or normalized values help reports and segments stay consistent across teams.
Protect reporting trust
CRM reports depend on the same data being used consistently across handoffs, lifecycle changes, and ownership changes.
- Blank, conflicting, or auto-filled stage values can make handoff reporting look better than the real process.
- Bulk enrichment should be tested on a sample before overwriting fields across thousands of records.
- Dashboards should be treated as a signal to inspect process and data quality, not as proof that the underlying records are clean.