CRM Data Cleanup Automation AI Prompt
Marketing AutomationintermediateClaude 3.5 Sonnet or GPT-4o. Both excel at creating structured, phased implementation plans. Claude is slightly better at handling complex conditional logic for data matching rules. GPT-4o provides more detailed automation workflow configurations. Choose Claude if you need detailed reasoning about data quality decisions; choose GPT-4o if you need specific CRM platform syntax.
When to Use This Prompt
Use this prompt when your CRM has accumulated data quality issues that impact campaign targeting, reporting accuracy, or sales productivity. It's essential before major marketing initiatives, system migrations, or when you notice declining email deliverability or segmentation problems.
The Prompt
You are a CRM data quality specialist. I need help creating a systematic plan to clean and standardize our CRM database. Our CRM contains [NUMBER] contacts across [DEPARTMENTS/REGIONS] with data quality issues including duplicate records, incomplete fields, inconsistent formatting, and outdated information.
## Current Data Quality Issues
- Duplicate contacts: [DESCRIBE DUPLICATION PATTERNS]
- Missing critical fields: [LIST FIELDS WITH >X% MISSING DATA]
- Formatting inconsistencies: [EXAMPLES: phone numbers, company names, address formats]
- Outdated records: [DESCRIBE AGE/STALENESS OF DATA]
- Data entry errors: [EXAMPLES: typos, wrong field usage]
## Cleanup Priorities
Rank these cleanup tasks by impact and feasibility:
1. Remove/merge duplicate records
2. Standardize phone number and email formats
3. Complete missing [CRITICAL FIELD NAMES]
4. Fix [SPECIFIC FORMATTING ISSUES]
5. Flag and archive inactive records
## Constraints
- Timeline: [WEEKS/MONTHS AVAILABLE]
- Team capacity: [NUMBER OF PEOPLE, SKILL LEVEL]
- Tools available: [CRM PLATFORM, AUTOMATION TOOLS]
- Risk tolerance: [HIGH/MEDIUM/LOW - IMPACT ON OPERATIONS]
Provide a phased cleanup roadmap including:
1. Pre-cleanup audit recommendations
2. Step-by-step cleanup sequence with specific rules for each phase
3. Validation checkpoints and quality metrics
4. Automation opportunities using [CRM PLATFORM] workflows
5. Team roles and responsibilities
6. Success metrics and monitoring dashboard setup
7. Prevention strategies for future data degradation
Include specific formulas, field mapping rules, and workflow configurations where applicable.
Get the Full AI Marketing Learning Path
Courses, workshops, frameworks, daily intelligence, and 6 proprietary tools — built for marketing leaders adopting AI.
Trusted by 10,000+ Directors and CMOs.
Tips for Best Results
- 1.Customize the [BRACKETS] sections with your actual CRM platform name, team size, and specific data issues—generic answers are less actionable than those tailored to your situation.
- 2.Ask a follow-up prompt to generate specific validation queries or SQL-like rules for your CRM platform after getting the roadmap.
- 3.Request the AI create a simple one-page checklist from the roadmap to share with your team—it increases adoption and accountability.
- 4.Use the output to estimate resource needs and timeline, then ask the AI to compress the plan if your actual capacity is lower than assumed.
Example Output
## CRM Cleanup Roadmap: 12-Week Implementation
### Phase 1: Audit & Preparation (Week 1-2)
**Pre-cleanup tasks:**
- Export full database backup to secure location
- Run duplicate detection report using [CRM Platform] built-in tools
- Identify records with >50% missing critical fields
- Document current data quality metrics baseline
- Create data cleanup validation rules
**Success metrics:** Baseline report completed, team trained on cleanup procedures
### Phase 2: Duplicate Resolution (Week 3-5)
**Approach:** Use CRM's merge functionality with these rules:
- Match on email address first (highest confidence)
- Secondary match: phone + last name
- Tertiary match: company + domain
- Merge strategy: Keep most recent activity record, consolidate interaction history
**Automation:** Set up workflow to flag new duplicates on entry
**Validation:** Manual review of 5% sample before final merge
### Phase 3: Standardization (Week 6-8)
**Field-by-field rules:**
- Phone: Convert to +1 (XXX) XXX-XXXX format
- Email: Lowercase all, remove spaces
- Company: Title case, remove "Inc." variations
- State: Convert to 2-letter abbreviations
**Automation:** Create bulk update workflows for each field
**Quality check:** Spot-check 100 records per field
### Phase 4: Enrichment & Validation (Week 9-11)
- Identify records with missing job title, company, or phone
- Prioritize by engagement level and lead score
- Use [ENRICHMENT TOOL] to fill critical gaps
- Manual review for high-value accounts
### Phase 5: Monitoring & Prevention (Week 12+)
- Implement data entry validation rules
- Create weekly data quality dashboard
- Establish monthly cleanup cadence
- Train team on proper data entry standards
Related Prompts
Related Reading
Get the Full AI Marketing Learning Path
Courses, workshops, frameworks, daily intelligence, and 6 proprietary tools — built for marketing leaders adopting AI.
Trusted by 10,000+ Directors and CMOs.