Why CRM data quality is the #1 ROI killer
Forrester estimates that bad data costs companies 20-30% of revenue — through misdirected marketing, lost opportunities, and wasted rep time. In Salesforce, dirty data specifically destroys: forecast accuracy (wrong close dates → wrong forecast), Einstein AI grounding (AI trained on bad data gives bad suggestions), reporting (dashboards no one trusts), and automation (Flow triggers on wrong field values). This 7-pillar framework is the data governance model we deploy for every Salesforce implementation.
Pillar 1: Completeness
Are required fields filled? Measure: % of records with all required fields populated. Target: >95%. Enforce via: validation rules (block save if required field is empty), page layouts (required fields at top), and Einstein Activity Capture (auto-fill activities). Run a weekly report showing completeness by rep — public accountability drives behavior.
Pillar 2: Accuracy
Is the data correct? Measure: % of records verified against an external source (ZoomInfo for contacts, Dun & Bradstreet for accounts). Target: >85%. Enforce via: data enrichment tools (ZoomInfo, Clearbit, D&B Optimizer), rep verification prompts (annual account review), and email bounce monitoring (auto-flag contacts with >3 bounces).
Pillar 3: Consistency
Is the data consistent across objects? Measure: % of accounts where billing address = shipping address (where applicable), % of opportunities where stage = forecast category. Target: >90%. Enforce via: workflow rules that sync fields across objects, formula fields (not manual entry), and picklist standardization (one list of values, enforced via validation rules).
Pillar 4: Timeliness
Is the data current? Measure: % of opportunities updated in the last 30 days, % of contacts with a last-activity date <90 days. Target: >80%. Enforce via: stale opportunity reports (emailed to managers weekly), Einstein Activity Capture (auto-stamp last activity), and escalation rules (opportunities with no activity for 30 days → manager review).
Pillar 5: Uniqueness
Are there duplicates? Measure: duplicate rate on accounts, contacts, leads (exact + fuzzy match). Target: <2%. Enforce via: duplicate rules (block on create, alert on edit), matching rules (fuzzy match on name+domain+phone), and quarterly dedup runs with Validity DemandTools or Salesforce's native duplicate management.
Pillar 6: Validity
Does the data follow business rules? Measure: % of email fields with valid format, % of phone numbers in standard format, % of opportunity amounts >$0. Target: >98%. Enforce via: regex validation rules on email/phone fields, picklist enforcement (no free-text industry), and flow validation (opportunity amount must be >0 to advance to Closed Won).
Pillar 7: Conformity
Does the data conform to data model standards? Measure: % of records using standard fields (not custom duplicates), % of picklist values used (>0 times). Target: >95%. Enforce via: data dictionary (document every field's purpose, owner, valid values), quarterly field audit (archive unused fields), and AppExchange tools (Salesforce Optimizer, Field Trip) that report field usage rates.
"Bad data costs companies 20-30% of revenue. In Salesforce, it specifically destroys forecasting, AI grounding, and reporting trust."
Key Takeaway
Measure 7 dimensions weekly: completeness (>95%), accuracy (>85%), consistency (>90%), timeliness (>80%), uniqueness (<2% dupes), validity (>98%), conformity (>95%). Use validation rules, duplicate rules, enrichment tools, and quarterly audits to maintain.