CRM Deduplication - Project Details / Task List
Tag: crm-dedup
Total Hours: 30h
Structure: Single Milestone (<=50h)
Milestone: CRM Deduplication
Description: A systematic data quality project that identifies, analyzes, and merges duplicate records within the CRM to ensure a single, accurate source of truth for every customer and prospect, establishing both initial cleanup and ongoing governance processes.
Task List: (CRM Deduplication) 1. Assessment & Configuration
Contains: Parts 1-2
| Task | Est | Description |
|---|---|---|
| 1. Audit Current Duplicate Landscape | 2h | Export sample data and quantify the duplicate problem across CRM objects. End state: Documented baseline of duplicate rates and patterns for contacts, leads, and accounts. • Pull export of contacts, leads, and accounts from CRM (last 90 days of created records minimum) • Run duplicate detection report using native CRM tools or trial of deduplication tool • Categorize duplicates by type: exact email matches, fuzzy name matches, company variations • Calculate duplicate percentage for each object (contacts typically 20-30% in uncleaned databases) • Identify top 5 sources creating duplicates (web forms, imports, manual creation, integrations) • Document sample duplicate clusters showing the merge complexity |
| 2. Interview Stakeholders on Pain Points | 1.5h | Gather qualitative input from CRM users on duplicate-related friction. End state: List of prioritized pain points and business rules that must be preserved during merge. • Interview CRM admin to understand current data entry processes and known problem areas • Interview 2-3 sales reps on "I'm already working that lead" frequency and manual workarounds • Interview marketing ops on email bounce rates and campaign segmentation issues • Document any custom fields or integration IDs that must be preserved during merges • Identify any existing duplicate handling processes (even informal ones) |
| 3. Define Matching Criteria and Master Record Rules | 2h | Establish the technical rules for what constitutes a duplicate and which record wins. End state: Documented deduplication rules approved by stakeholders. • Define primary matching criteria (email address is typically the anchor for B2B) • Define secondary matching criteria (phone, company name + first/last name combination) • Establish fuzzy matching thresholds (how similar must names be to flag as potential duplicates) • Create master record selection hierarchy (e.g., most recent activity > most complete data > oldest created date) • Document special handling for free email domains (Gmail, Hotmail should not be master unless only option) • Get sign-off from RevOps lead and CRM admin on rules before proceeding |
| 4. Evaluate Tool Options Against Requirements | 2.5h | Compare deduplication tool options based on client's CRM, budget, and matching needs. End state: Tool selected with procurement initiated. • Document client's CRM (Salesforce or HubSpot) and any integrations that affect deduplication • Evaluate native CRM tools (Salesforce Duplicate Management, HubSpot native deduplication) • Evaluate third-party options: Insycle, Dedupely, Cloudingo, DemandTools, RingLead • Compare on: fuzzy matching capabilities, preview/undo options, bulk processing limits, cost per user • Consider integration limitations (e.g., HubSpot can't merge companies when Salesforce integration is active) • Present recommendation with pros/cons to client and get budget approval |
| 5. Configure Tool with Matching Rules | 3h | Set up the deduplication tool with the approved matching criteria and merge rules. End state: Tool configured and ready for test run. • Install/connect deduplication tool to CRM environment • Configure primary matching field (email) with exact match rules • Configure secondary matching fields with appropriate fuzzy match thresholds • Set up master record selection logic per approved hierarchy • Configure field merge behavior (which fields take master value vs. concatenate vs. most recent) • Define handling for related records (activities, tasks, opportunities linked to merged records) • Save configuration and document settings for future reference |
Task List: (CRM Deduplication) 2. Execution, Prevention & Handoff
Contains: Parts 3-5
| Task | Est | Description |
|---|---|---|
| 6. Run Test Batch and Validate Accuracy | 2h | Process a small test batch to validate matching accuracy before full execution. End state: Validated deduplication rules with acceptable false positive/negative rates. • Select test batch of 300-500 records across objects (contacts, leads, accounts) • Run deduplication tool in preview mode (no actual merges) • Manually review 50+ potential duplicate pairs for accuracy • Calculate false positive rate (records flagged that are NOT duplicates) • Calculate false negative rate (known duplicates that were missed) • Adjust matching thresholds if false positive rate exceeds 5% • Document any edge cases requiring special handling |
| 7. Execute Full Deduplication in Batches | 3.5h | Process all duplicate records in controlled batches with rollback capability. End state: All identified duplicates merged with documented results. • Ensure CRM backup is completed before proceeding (verify with client IT/admin) • Start with lowest-risk object (typically contacts before accounts) • Process in batches of 500-1000 records to maintain control • Review batch results before proceeding to next batch • Document merge counts: records before, records after, merge ratio • Handle exceptions and edge cases flagged during batch processing • Move to next object type after completing current object • Generate final deduplication report with total records merged by object |
| 8. Verify Data Integrity Post-Merge | 1.5h | Confirm that merges preserved critical data and related records remained linked. End state: Verified clean database with no data loss or broken relationships. • Spot-check 20-30 merged records to verify field values preserved correctly • Verify related records (activities, opportunities, tasks) are still linked to surviving records • Check that integration IDs (Salesforce ID, HubSpot ID, etc.) were preserved per rules • Run critical reports/dashboards to confirm data integrity • Address any data issues discovered during verification |
| 9. Set Up Real-Time Duplicate Alerts | 2h | Configure CRM to warn users when creating potential duplicates. End state: Users receive alerts before creating duplicate records. • Enable native duplicate detection rules in CRM (Salesforce Matching Rules, HubSpot duplicate management) • Configure alert behavior (block vs. warn) based on match confidence • Set up duplicate alerts for key entry points: manual creation, lead forms, imports • Configure alert message with guidance on how to find existing record • Test duplicate alert with sample records across each entry point |
| 10. Create Validation Rules to Prevent Duplicates | 2h | Implement hard blocks on duplicate creation where appropriate. End state: System-enforced prevention of obvious duplicates. • Create validation rule to require email address on contact/lead creation (prevents duplicates impossible to match) • Configure required field validation to ensure matching criteria fields are populated • Set up email domain validation (standardize formatting, lowercase) • Create workflow to standardize company names at entry (trim whitespace, standardize abbreviations) • Test validation rules to ensure they don't block legitimate record creation |
| 11. Build Ongoing Automated Scans | 2h | Configure recurring deduplication scans to catch duplicates that slip through prevention. End state: Scheduled automated duplicate detection running weekly/monthly. • Schedule recurring duplicate scan (weekly for high-volume databases, monthly for lower volume) • Configure scan to generate report of potential duplicates for review • Set up notification to RevOps admin when scan completes • Define SLA for reviewing and resolving flagged duplicates • Create process for handling low-confidence matches requiring manual review |
| 12. Create Data Governance Documentation | 1.5h | Document all deduplication rules, processes, and procedures for ongoing maintenance. End state: Complete Data Governance SOP ready for client ownership. • Document matching criteria and master record selection rules • Create duplicate handling procedure (what to do when duplicate is found) • Document import best practices to prevent duplicate creation • Create troubleshooting guide for common duplicate scenarios • Document tool admin access and configuration settings • Include escalation path for complex duplicate situations |
| 13. Train Team on Data Entry Best Practices | 1.5h | Conduct training session with end users on preventing duplicates at the source. End state: Team trained and aligned on duplicate prevention. • Schedule 30-45 minute training session with sales and marketing teams • Cover: how duplicates impact their work, what prevention is now in place, how to respond to alerts • Demonstrate how to search for existing records before creating new ones • Review data entry standards (email format, company name standardization) • Provide quick-reference card with duplicate prevention tips • Record session for onboarding new team members |
| 14. Hand Off to Client RevOps Team | 3h | Transfer ownership of deduplication tools and ongoing maintenance to client. End state: Client self-sufficient with admin access, documentation, and scheduled reviews. • Transfer deduplication tool admin credentials to client RevOps team • Walk through tool configuration and how to adjust matching rules • Review scheduled scan process and how to interpret results • Deliver complete documentation package • Schedule 30-day check-in to review new duplicate creation rate • Provide guidance on when to re-run bulk deduplication (typically quarterly) |
Summary
- Total Task Lists: 2 (consolidated from 5 Parts)
- Total Tasks: 14 (one per Step)
- Total Hours: 30h
- Generated from: playbook_crm-deduplication.md
- Generated on: 2025-12-31