CRM Deduplication Ongoing Tool — Implementation

Project One-Pager

CRM Deduplication Ongoing Tool One-Pager

Project Type

• Category: Technical
• Primary Deliverable: Deployed and configured deduplication tool with automated matching rules, prevention logic, merge workflows, and ongoing monitoring dashboard

Phase Relevance

Phase	Applies?	Weight	Notes
1. Strategy	Yes	Medium	Tool evaluation, requirements gathering, rule design
2. Engineering	Yes	Heavy	Tool setup, matching rules, automation, prevention logic
3. Enablement	Yes	Medium	Training on tool usage, manual review queue, maintenance
4. Handoff	Yes	Medium	Documentation package, monitoring cadence, ownership

· · ·

Phase Overview

  ┌──────────────┐     ┌──────────────┐     ┌──────────────┐     ┌──────────────┐
  │ 1. STRATEGY  │────▶│ 2. ENGINEER  │────▶│3. ENABLEMENT │────▶│  4. HANDOFF  │
  │    Medium    │     │    Heavy     │     │    Medium    │     │    Medium    │
  │ 1a→1b→1c→1d │     │ 2a→2b→2c→2d │     │ 3a→3b→3c→3d │     │ 4a→4b→4c→4d │
  └──────────────┘     └──────────────┘     └──────────────┘     └──────────────┘
   Tool evaluation      Tool setup +         Training + review    Docs + monitoring
   + rule design        automation build     queue + maintenance  cadence + ownership

This project's flow:

• Full 4-phase. Medium strategy (tool evaluation + rule design), heavy engineering (tool configuration + automation), standard enablement and handoff.
• Phase 1 includes tool selection and procurement — budget approval can be a bottleneck. Plan for 1-2 week procurement lead time after selection.
• Phase 2 is the core of the work: configuring matching rules for Contacts/Leads and Accounts/Companies, defining master record survivorship, building prevention and automation workflows, and testing.
• No phases skipped. All clients need training (Phase 3) to manage manual review queues and ongoing rule tuning.

· · ·

Pre-Kickoff (1a)

Track A: Customer Homework

• Watch intro video explaining deduplication tool project scope and why ongoing prevention matters
• Complete intake form: CRM platform, edition/tier, current duplicate volume estimates, integration landscape
• Provide list of critical fields that must be preserved during merges
• Confirm budget approval range for tool licensing
• Identify RevOps owner for ongoing tool management
• Provide CRM admin access credentials and API permissions

Track B: Architect Prep

• Run duplicate audit reports in CRM (Contacts, Leads, Accounts/Companies)
• Document duplicate sources: manual entry, imports, form submissions, integrations
• Quantify duplicate rate per object (e.g., "15% of Contacts have duplicates")
• Review existing deduplication rules or validation checks
• Assess CRM-MAP sync configuration (HubSpot-Salesforce sync, if applicable)
• Prepare tool comparison matrix with 2-3 shortlisted options

· · ·

Refinement Loop (1b → 1c → 1d)

Meeting	Sub-Phase	Focus	Stakeholder	Output
Kickoff	1b	Present duplicate audit, review tool options, gather rules	RevOps, Sales Ops	Requirements confirmed for v1
Refinement 1	1c	Review tool recommendation, define matching rules, discuss master record criteria	RevOps, IT/Admin	Tool selected, rules drafted
Refinement 2	1c	Finalize survivorship logic, prevention strategy, automation thresholds	RevOps, Compliance	Configuration spec finalized
Sign-Off	1d	Approve tool, rules, thresholds, and automation schedule	All stakeholders	Strategic sign-off, procurement

· · ·

Phase Checklists

Phase 1: Strategy

• 1a. Pre-Kickoff complete (Track A + Track B)
• 1b. Kickoff call held — duplicate audit presented, tool options reviewed
• 1c. Refinement loop complete — tool selected, rules designed, thresholds set
• 1d. Strategic sign-off obtained — budget approved, procurement initiated

Phase 2: Engineering

• 2a. Tech spec created — matching rules, survivorship logic, automation schedules
• 2b. Engineering handoff meeting held — tool admin walkthrough
• 2c. Build complete — tool connected, rules configured, automation live
• 2d. QA/Test + customer sign-off — preview merge run approved

Phase 3: Enablement

• 3a. Training materials prepped — manual review SOP, quick-reference guide
• 3b. Training sessions delivered — RevOps team trained on tool usage
• 3c. Hypercare period complete — 30-day monitoring, queue review support
• 3d. Enablement sign-off — team operating independently

Phase 4: Handoff

• 4a. Maintenance schedule documented and handed off
• 4b. Internal handoff (SME → Architect) complete
• 4c. External handoff (delivery team → Customer) complete
• 4d. Project closed and archived

· · ·

Document Types

Working Documents (iterate together)

Document	Purpose	When Complete
Duplicate Audit Report	Baseline: volume, sources, patterns per object	All objects audited, rates quantified
Tool Comparison Matrix	Evaluate 2-3 tools against requirements	Client selects tool, budget approved
Matching Rules Spec	Document matching fields, thresholds, cross-obj	Rules validated with sample data
Survivorship Rules Doc	Master record selection + field-level rules	Compliance reviewed, opt-out handling confirmed

Deliverables (polished outputs)

Deliverable	Created From	Customer Uses For
Tool Configuration Runbook	Matching + survivorship	Reference for rule changes and maintenance
Automation Schedule Doc	Build outputs	Understanding when auto-merge runs
Manual Review SOP	Training materials	Day-to-day duplicate review process

· · ·

Enablement Details

Training Types

Type	Audience	Focus	Duration
Leadership	VP Sales, VP Marketing	Why dedup matters, dashboard metrics, ROI review	20 min
Technical	RevOps, Sales Ops, CRM Admin	Tool usage, manual review queue, rule tuning, unmerge process	45 min

Hypercare

• Applies: Yes
• Duration: 30 days
• Office Hours: Yes — weekly 30-min slot for queue review and rule tuning questions

Training Assets to Create

• Video walkthrough: Tool dashboard and manual review queue
• Video walkthrough: How to run on-demand deduplication scans
• Doc: Manual review SOP with merge/reject criteria
• Doc: Quick-reference guide for common tasks (run scan, review queue, adjust threshold, unmerge)

· · ·

Handoff & Retention

Internal Handoff (SME → Architect)

• Key context for Architect: Tool name and version, automation schedule, current duplicate rate, manual review queue cadence
• Escalation trigger: Any matching rule changes, survivorship logic updates, cross-object rule modifications, or new integration data sources

External Handoff (Delivery Team → Customer)

• Final meeting agenda: Review delivered configuration, walk through maintenance schedule, demo unmerge/rollback, confirm ongoing support path
• Documentation package: Configuration runbook, automation schedule, manual review SOP, training video walkthroughs, quick-reference guide

Maintenance Schedule

• Monthly: Review manual queue, check duplicate rate trending, verify automation runs
• Quarterly: Review matching rule accuracy, adjust thresholds, audit new data sources
• Who owns: Single Project = customer owns | Dedicated = Architect owns

Retention/Expansion Path

If Single Project: Upsell: Managed Services → if no → Downsell: Another project (CRM Data Enrichment, Lead Routing) → Retry retainer

If Multi-Project (Dedicated):

• Refinement check-in scheduled: ~1 quarter out
• Internal prep trigger: 2 weeks before
• Decision: Architect handles / SME needed (rule changes = SME)

· · ·

Key Assets

Asset	When Used
Duplicate Audit Template	Phase 1a (Architect Prep)
Tool Comparison Matrix	Phase 1b-1c
Matching Rules Spec	Phase 2a-2c
Manual Review SOP	Phase 3b, Phase 4c
Configuration Runbook	Phase 4c

· · ·

Definition Alignment Terms

Term	Typical Definition
Duplicate	Two or more records representing the same real-world entity (person or company) in the CRM
Master Record	The surviving record after a merge — retains the record ID and becomes the single source of truth
Survivorship Rules	Logic determining which field values are kept from merged records (e.g., most recent, most complete)
Matching Rule	Criteria used to identify potential duplicates (e.g., exact email match, fuzzy name match)
Confidence Score	Numeric or categorical measure of how likely two records are true duplicates (drives auto-merge vs. review)
False Positive	Two records flagged as duplicates that are actually distinct entities
False Negative	Two records that are true duplicates but were not detected by matching rules
Cross-Object Matching	Deduplication across different objects (e.g., Contact vs. Lead) rather than within the same object
Fuzzy Matching	Matching algorithm that accounts for name variations, typos, abbreviations (e.g., "Inc" vs "Incorporated")
Manual Review Queue	Holding area for potential duplicates below the auto-merge confidence threshold, requiring human decision

· · ·

Common Gotchas

• HubSpot auto-creates Company records from Contact activity — disable this setting before configuring dedup rules if Salesforce is the source of truth, or duplicates re-enter constantly [1]
• Salesforce limits you to 5 active Duplicate Rules per object and 5 active Matching Rules per object — plan rule allocation carefully [2]
• Merging records with active Opportunities or open Cases can break ownership and workflow triggers — test merge behavior on records with related objects before bulk operations
• Opt-out/unsubscribe preferences must be preserved during merge — configure survivorship to always keep the most restrictive opt-out status from any merged record (GDPR/CAN-SPAM compliance)
• Third-party tools connected via API may re-sync deleted duplicates back into the CRM if sync rules are not updated to respect the merge

· · ·

Methodology Options (if applicable)

Option	When to Use	Complexity
Native CRM Tools Only	Low duplicate volume (<5%), simple matching needs, tight budget	Low
Third-Party Tool (Insycle/Dedupely)	Mid-market, moderate volume, HubSpot or Salesforce, need automation	Medium
Enterprise Platform (Openprise/RingLead)	High volume, complex matching, multi-CRM, cross-object needs	High

See Methodology for detailed tool evaluation frameworks, matching algorithm comparisons, and scoring rubrics.

---

Phase 1: Strategy

Goal: Assess duplicate landscape, select deduplication tool, design matching and survivorship rules, and get stakeholder sign-off before building.

Output: Approved tool selection + Configuration Spec (matching rules, survivorship logic, automation schedule, prevention strategy) signed off by RevOps and compliance stakeholders.

1a. Pre-Kickoff

Two parallel tracks run before the kickoff call.

Track A: Customer Homework

What we send:

Item	Purpose	Format
Intro video	Explain the project: why ongoing dedup matters, what the tool does	Video (5-10 min)
Definition Alignment Document	Get sign-off on terms: duplicate, master record, survivorship, matching	Google Doc
Pre-filled intake form	CRM platform, edition, integrations, current dedup pain, budget range	Google Form or Doc

Specific intake questions:

• CRM platform and edition (Salesforce Enterprise, HubSpot Professional, etc.)
• Current estimated duplicate rate (or "unknown")
• Primary duplicate sources: manual entry, imports, form submissions, integrations
• Existing dedup rules or validation checks in place
• CRM-MAP sync configuration (HubSpot-Salesforce sync, Marketo sync, etc.)
• Critical fields that must be preserved during merges
• Budget range for tool licensing ($5K-$50K/year typical range)
• Identified RevOps owner for ongoing management

Completion tracking: Follow up to ensure completion. Do not cancel kickoff if incomplete, but push hard after for admin access and budget range — these gate the tool selection timeline.

Track B: Architect Prep

What the Architect does:

Step	Action	Output
1	Run duplicate audit reports in CRM for all target objects	Duplicate rate per object, volume, pattern analysis
2	Document duplicate sources (manual entry, imports, forms, integrations)	Source breakdown with estimated contribution per channel
3	Review CRM-MAP sync configuration	Sync direction, conflict resolution rules, auto-create settings
4	Research tool options based on CRM platform and requirements	Tool comparison matrix (2-3 options)
5	Create kickoff assets: audit summary, tool comparison, questions list	Kickoff presentation deck

Critical: Mark all audit findings as ASSUMED until validated with the RevOps team. They may know about duplicate sources that cannot be detected from CRM data alone.

Stakeholder Alignment Document

Get stakeholder sign-off on terms BEFORE configuring anything.

Term	Our Definition	Internally Approved?
Duplicate	Two or more records representing the same real-world entity in the CRM	[ ] Yes / [ ] No
Master Record	The surviving record after merge — retains record ID, becomes source of truth	[ ] Yes / [ ] No
Survivorship Rules	Logic for which field values are kept from merged records	[ ] Yes / [ ] No
Confidence Threshold	The score above which duplicates auto-merge vs. route to manual review	[ ] Yes / [ ] No
Cross-Object Match	Deduplication across Contacts and Leads (not just within one object)	[ ] Yes / [ ] No

Instructions to customer:

Review each definition with your RevOps and compliance teams. Check "Yes" when approved. We cannot configure matching rules until terms are aligned — particularly "master record" selection criteria and "survivorship" logic, as these are irreversible once merges execute.

---

1b. Kickoff Call

Purpose: Present duplicate audit findings, review tool options, and gather matching rule requirements. We walk in with the audit done — customer reacts, not creates from scratch.

Agenda (60-90 min)

Time	Topic	What Happens
0-20	Present duplicate audit	"Here's what we found: X% duplicate Contacts, Y% Accounts"
20-35	Validate sources and patterns	Confirm which sources generate the most duplicates
35-50	Review tool options	Walk through comparison matrix, discuss pros/cons
50-65	Matching rule priorities	Which fields matter most? Email, domain, name? Cross-object?
65-75	Master record preferences	Most complete? Most recent? Earliest created? Field-by-field?
75-90	Next steps	Homework: budget approval, admin access, schedule refinement

What We Bring

• Duplicate audit summary (rates per object, source breakdown, pattern examples)
• Tool comparison matrix with 2-3 options
• Draft matching field priority list (based on B2B best practices)
• Definition Alignment Document (pre-filled with recommendations)

What We Leave With

• Validated duplicate sources and patterns
• Tool preference or shortlist narrowed to 1-2
• Matching field priorities confirmed
• Master record selection preference captured
• Homework assigned: budget approval, admin access provisioning

---

1c. Alignment Loop & Strategic Meeting Cadence

Purpose: Iterate on tool selection, matching rules, and survivorship logic until sign-off.

The Pattern

Kickoff Call (gather requirements, present audit)
    |
    v
Architect drafts matching rules spec, refines tool recommendation --> v1
    |
    v
Meeting 2 (review rules, finalize tool selection) --> Architect refines --> v2
    |
    v
Meeting 3 (survivorship logic, prevention strategy, automation schedule) --> Architect finalizes --> v3
    |
    v
Final Review --> Sign-off

Before Each Meeting

1. Update matching rules spec with previous meeting's feedback
2. Research any open questions (tool capabilities, CRM limitations, compliance requirements)
3. Prepare decision items for next validation round

During Each Meeting

1. Walk through current version of configuration spec
2. Capture corrections and refinements
3. Validate what is now CONFIRMED vs. still ASSUMED
4. Identify remaining decisions needed

After Each Meeting

1. Update configuration spec (matching rules, survivorship, prevention, automation)
2. Track what moved from ASSUMED to CONFIRMED
3. Update tool procurement status if applicable

Meeting Schedule

Meeting Type	Focus	Stakeholder
Kickoff	Audit results, tool options, initial requirements	RevOps Lead, Sales Ops
Tool Selection + Rules	Finalize tool, design matching rules per object	RevOps, IT/Admin
Prevention + Automation	Survivorship logic, prevention strategy, automation design	RevOps, Compliance (if needed)
Final Review	Full configuration spec walkthrough, sign-off	All stakeholders

Typical Timeline

Milestone	Timing
Pre-kickoff prep	2-3 days
Kickoff call	Day 1 of engagement
Meeting loop	1-2 weeks (tool procurement can extend by 1-2 weeks)
Final review + sign-off	When tool procured, rules confirmed, thresholds agreed

---

1d. Strategic Sign-Off

Purpose: Confirm we have everything before proceeding to build.

Validation Checkpoint

• Definition Alignment Document signed off by stakeholders
• Duplicate audit validated by RevOps team
• Tool selected and procurement completed (license active, admin access granted)
• Matching rules designed for Contacts/Leads (primary + secondary fields, fuzzy thresholds)
• Matching rules designed for Accounts/Companies (domain, name variations, parent/child)
• Cross-object matching scope confirmed (Contact vs. Lead dedup: yes/no)
• Master record survivorship rules defined (field-level, opt-out preservation)
• Confidence thresholds set (auto-merge cutoff vs. manual review cutoff)
• Prevention strategy agreed (real-time alerts, blocks, form/import rules)
• Automation schedule defined (daily/weekly, notification recipients)
• No blockers for engineering

Decision Point

• Proceed to Engineering — Tool procured, rules designed, thresholds set. This project always proceeds to Engineering.

---

Phase 2: Engineering

Goal: Configure the deduplication tool end-to-end: matching rules, survivorship, prevention, automation, and manual review queue. Test and validate with preview merge runs.

Output: Fully configured deduplication tool, tested against real data, initial duplicate backlog cleared, customer-approved.

Project Type	Engineering Weight	Notes
CRM Dedup Tool	Heavy (60-70%)	Core of the project — tool setup, rules, automation, testing

Sub-Phases

2a Tech Spec --> 2b Engineering Handoff --> 2c Build --> 2d Test

---

2a. Tech Spec

Purpose: Translate strategic configuration spec into technical implementation plan.

Input: Signed-off configuration spec from Phase 1 (matching rules, survivorship logic, thresholds, automation schedule, prevention strategy)

What happens:

1. Translate matching rule designs into tool-specific configurations (field mappings, algorithm selections, threshold values)
2. Map survivorship rules to tool settings (field-by-field merge behavior)
3. Document API connection requirements and permissions
4. Sequence the build order: connection first, then Contact/Lead rules, then Account/Company rules, then prevention, then automation, then manual review queue

Output: Technical implementation plan containing:

• Tool connection checklist (OAuth/API key, permissions, sandbox vs. production)
• Matching rule configurations per object (exact fields, fuzzy fields, algorithm type, thresholds)
• Cross-object matching configuration (Contact-to-Lead, if applicable)
• Survivorship rule mappings (field-level: keep most recent, keep most complete, keep non-null, always preserve opt-out)
• Prevention rule setup (real-time detection: warn, block, or auto-merge; form/import rules)
• Automation schedule configuration (frequency, confidence threshold, notification recipients)
• Manual review queue setup (routing rules, assignment, checklist criteria)
• Test plan (known duplicate pairs, known non-duplicate pairs, edge cases)

---

2b. Engineering Handoff

Purpose: Review tech specs before configuring the tool.

Agenda (30-45 min):

Time	Topic	What Happens
0-15	Walk through tech spec	Review matching rules, survivorship, prevention, automation
15-30	Clarify tool specifics	API permissions, CRM field access, sandbox availability
30-45	Confirm build sequence	Agree on order, identify risks (sync conflicts, API limits)

What to bring:

• Configuration spec (from Phase 1)
• Tech spec (from 2a)
• Tool admin credentials and CRM access confirmation

What to leave with:

• Approved tech spec with build sequence
• Known risks: HubSpot auto-create company setting, Salesforce duplicate rule limits, sync conflicts
• Test data: known duplicate pairs and known non-duplicate pairs for validation

---

2c. Build (Configure)

Purpose: Configure the deduplication tool in the client's CRM environment.

Input: Approved tech spec from 2b

Build sequence:

Step	Component	Details
1	Tool account + CRM connection	Create accounts, connect via OAuth/API, verify read/write permissions
2	Contact/Lead matching rules	Primary: email (exact). Secondary: name (fuzzy), phone, company, domain. Set thresholds.
3	Account/Company matching rules	Primary: domain/website (exact). Secondary: company name (fuzzy). Handle parent/child.
4	Cross-object matching (if applicable)	Contact-to-Lead matching rules, confidence thresholds
5	Master record survivorship rules	Field-level: most complete, most recent, non-null. Always preserve opt-out. Related records.
6	Real-time duplicate prevention	Alert/block on record creation. Form submission prevention. Import duplicate checking.
7	Automated merge workflows	Schedule (daily/weekly). High-confidence auto-merge. Lower-confidence to review queue.
8	Manual review queue	Assignment rules, review criteria checklist, unmerge capability confirmed
9	Dashboard/reporting	Duplicate rate trending, merge volume, queue backlog, false positive tracking

CRM-specific considerations:

Salesforce:

• Use native Duplicate Rules + Matching Rules for prevention (up to 5 active per object) [2]
• Third-party tool handles fuzzy matching and automation beyond native capabilities
• Test that Duplicate Rules do not conflict with third-party tool's real-time detection

HubSpot:

• Disable "auto-create company from contact activity" if Salesforce is source of truth [1]
• Native dedup requires Professional or Enterprise subscription; bulk dedup requires Operations Hub [3]
• Third-party tools (Insycle, Koalify, Dedupely) provide automation that HubSpot native lacks

Build tracking:

• Tool account created and CRM connected
• Contact/Lead matching rules configured
• Account/Company matching rules configured
• Cross-object matching configured (if applicable)
• Survivorship rules defined and configured
• Real-time prevention active
• Automated merge workflow scheduled
• Manual review queue operational
• Dashboard/reporting built

---

2d. QA / Test + Sign-Off

Purpose: Verify all configurations work correctly and get customer approval before production rollout.

Two types of testing:

Type	Who	Purpose
Technical Testing	Our team	Verify matching accuracy, false positive/negative rates, merge integrity
Customer Testing	Customer	Verify tool catches their known duplicates, doesn't merge non-duplicates

Technical testing checklist:

• Matching rules correctly identify known duplicate pairs (true positives)
• Matching rules do not flag known non-duplicate pairs (false positive rate acceptable — target <5%)
• Matching rules catch fuzzy matches (name variations, typos, abbreviations)
• Cross-object matching works correctly (Contact vs. Lead)
• Master record selection follows survivorship rules
• Field-level survivorship preserves correct values (most recent, most complete, non-null)
• Opt-out/unsubscribe preferences always preserved from any merged record
• Related records (activities, opportunities, notes) transfer correctly to master
• Real-time prevention fires on record creation (alerts/blocks as configured)
• Prevention works for form submissions and data imports
• Automated merge workflow runs on schedule with correct thresholds
• Low-confidence matches route to manual review queue
• Unmerge/rollback capability confirmed working
• Dashboard displays accurate metrics

Initial production run (preview mode):

1. Generate preview report of all detected duplicates across objects
2. Review preview with client RevOps team — flag any unexpected matches
3. Address false positives by adjusting rules or thresholds
4. Execute merge on approved duplicates (high confidence first, then reviewed lower confidence)
5. Verify post-merge: related records intact, no broken workflows, audit trail complete
6. Generate post-merge report: records merged, duplicate rate reduction, remaining queue

Engineering sign-off checkpoint:

• All matching rules tested and validated
• False positive rate below 5%
• Initial duplicate backlog cleared via approved preview merge
• Post-merge data integrity verified
• Automation running on schedule
• Customer has tested and approved results
• Ready for enablement

Decision point:

• Proceed to Enablement — Tool configured, tested, initial backlog cleared
• Loop back to Build — False positive rate too high, rules need adjustment

---

Phase 3: Enablement

Goal: RevOps team can independently operate the deduplication tool: manage the manual review queue, run on-demand scans, adjust rules, and troubleshoot.

Output: Trained team with documentation, stabilized tool running in production, no critical issues.

Sub-Phases

3a Training Prep --> 3b Training Sessions --> 3c Hypercare --> 3d Enablement Sign-Off

---

3a. Training Prep

Purpose: Create training materials from the configuration spec and build documentation.

Input: Configuration spec + tech spec + built/tested tool + post-merge report

What happens:

1. Create video walkthrough scripts (dashboard, review queue, on-demand scan)
2. Write manual review SOP (when to merge, when to reject, when to escalate)
3. Create quick-reference guide (common tasks: run scan, review queue, adjust threshold, unmerge)
4. Draft FAQ from common questions encountered during build and testing

Output: Training package containing:

• Video walkthrough scripts (3 videos: dashboard overview, manual review process, on-demand scanning + unmerge)
• Manual Review SOP (decision criteria, merge checklist, escalation triggers)
• Quick-reference guide (1-page task reference)
• FAQ document (10-15 common questions with answers)

---

3b. Training Sessions

Purpose: Transfer knowledge to the RevOps/Sales Ops team.

Two types of training:

Type	Audience	Focus
Leadership training	VP Sales, VP Marketing	Why dedup matters: duplicate records waste 27% of rep selling time [4]. Dashboard metrics, ROI tracking.
Technical handoff	RevOps, Sales Ops, CRM Admin	Manual review queue operation, on-demand scanning, rule adjustment, unmerge

Technical training session (45 min):

1. Dashboard walkthrough: duplicate rate trending, merge volume, queue backlog
2. Manual review queue: how to review, merge, reject, and escalate matches
3. On-demand scanning: how to run a scan outside the automated schedule
4. Rule adjustment: when and how to modify matching thresholds
5. Unmerge/rollback: how to reverse an incorrect merge
6. Automation monitoring: how to verify scheduled runs are executing

Leadership training session (20 min):

1. Dashboard metrics overview: duplicate rate, reduction trend, queue health
2. Business impact: 10-30% of CRM records are typically duplicates, costing $12.9M annually in poor data quality for the average organization [5][6]
3. Success criteria: maintain duplicate rate below 5% ongoing, queue processed weekly

Training delivery:

1. Schedule sessions with appropriate stakeholders
2. Deliver live training (record for future reference)
3. Record supplementary video walkthroughs for async consumption
4. Answer questions, note gaps for FAQ updates

Output:

• Trained RevOps team
• Video walkthrough recordings (3 videos)
• Updated FAQ based on training questions

---

3c. Hypercare

Purpose: Intensive post-launch support to stabilize the tool and build team confidence.

Duration: 30 days

What happens:

• Weekly 30-minute office hours: RevOps team can ask questions, review queue together, troubleshoot
• Monitor automated merge runs: verify no errors, check false positive reports
• Review and adjust matching thresholds if false positive or negative rates shift
• Support the team through their first full manual review queue cycle
• Address any new duplicate sources discovered (new integration, new form, new import process)

What is in scope:

• Threshold adjustments based on production data
• Queue review support and merge decision guidance
• Minor rule modifications (adding a secondary matching field, adjusting fuzzy tolerance)

What is out of scope:

• Adding entirely new objects to dedup (requires scoping as new project)
• Changing the dedup tool (requires new procurement cycle)
• CRM-MAP sync reconfiguration (separate project)

When to skip: Never — every CRM deduplication tool deployment needs a hypercare period. Production data always surfaces edge cases that testing does not.

Output: Stabilized tool, no critical issues, team processing manual review queue independently

---

3d. Enablement Sign-Off

Purpose: Confirm customer team can operate the dedup tool independently.

Validation checkpoint:

• All training sessions delivered
• Training recordings and documentation provided
• Hypercare period complete (30 days)
• Manual review queue being processed on schedule by RevOps team
• No critical issues outstanding
• Automated merges running without errors
• Team can run on-demand scans without support
• Team understands when to escalate
• Ready for handoff

Decision point:

• Proceed to Handoff — Team is enabled, tool stabilized
• Extend Hypercare — Queue backlog growing, team not yet confident, rule tuning still needed

---

Phase 4: Handoff

Goal: Clean project close with maintenance plan established and retention/expansion path set.

Output: Maintenance schedule documented, internal context transferred, customer owns the tool, project archived, future revenue path established.

Structure:

4a Maintenance Schedule --> 4b Internal Handoff --> 4c External Handoff --> 4d Project Close
                          (SME --> Architect)       (Team --> Customer)    (Archive + Debrief)

Maintenance ownership by engagement type:

Engagement Type	Who Owns Maintenance	Handed Off At
Single Project	Customer owns	4c (External Handoff) — customer receives maintenance schedule and runs it themselves
Dedicated (Multi-Project)	Architect owns	4b (Internal Handoff) — Architect receives maintenance schedule and runs it for customer

---

4a. Maintenance Schedule

Purpose: Document what needs ongoing attention to keep the dedup tool effective and duplicate rates low.

Standard Maintenance Framework

Monthly Tasks:

Monthly Task	What to Check	Red Flag Threshold
Manual Review Queue	Queue backlog size, processing frequency	>50 unreviewed matches (queue growing, not shrinking)
Duplicate Rate Trending	Current duplicate rate vs. previous month	Rate increasing >2% month-over-month
Automation Run Verification	Scheduled merges executing, no errors in logs	Any failed run or error notification
New Data Source Check	Any new integrations, forms, or import processes added	New source without prevention rules configured

Quarterly Tasks:

Quarterly Task	What to Review	Action if Off-Track
Matching Rule Accuracy Audit	False positive and false negative rates from review queue	Adjust thresholds or add/remove matching fields
Survivorship Rule Review	Are the right field values surviving merges?	Update survivorship logic if business rules changed
Cross-Object Match Effectiveness	Contact-to-Lead matching catching real duplicates?	Adjust cross-object thresholds or criteria
Tool License Review	Usage vs. license limits, feature utilization	Downgrade/upgrade if usage patterns changed

After First Business Cycle (30-90 days post-launch):

• Duplicate rate reduction: target 80%+ reduction from baseline within 90 days
• Ongoing duplicate rate: should be below 5% and trending toward 2% [6]
• Manual review queue velocity: team processing queue within 1 week of matches appearing
• False positive rate: below 5% of flagged matches

Refinement Triggers (when to re-engage):

Trigger	Threshold	Response
Duplicate rate climbing	>5% for 2+ consecutive months	Investigate new sources, adjust prevention rules
False positive rate spiking	>10% of flagged matches are non-dupes	Re-tune matching thresholds, review fuzzy tolerance
Queue backlog growing	>100 unreviewed matches for 30+ days	Re-train team, adjust auto-merge threshold upward
New CRM integration added	Any new data source connected to CRM	Configure prevention rules for new source immediately

Every 6-12 Months:

• Full matching rule recalibration: test rules against fresh data sample, update thresholds
• Tool version/feature review: evaluate new capabilities released by vendor
• Compliance review: verify opt-out preservation rules still meet current regulations
• Duplicate source analysis: reassess which entry points generate the most duplicates

---

4b. Internal Handoff (SME → Architect)

Purpose: Transfer context so the Architect can manage the ongoing relationship.

What the Architect needs to know:

• Tool name, version, and login credentials (admin vs. read-only)
• Automation schedule (what runs when, who gets notified)
• Current duplicate rate and trend direction
• Manual review queue cadence and who processes it
• Common issues and how to resolve (see troubleshooting guide below)
• When to escalate back to SME

Escalation guidelines:

Issue Type	Who Handles	Example
Queue review questions, minor threshold adjustments	Architect	"Should we merge these two?" "Queue is backed up."
Rule changes, new object dedup, survivorship changes	SME	"We need to dedup Deals now." "Matching rules wrong."

Troubleshooting Guide:

Scenario	Resolution
Automated merge run failed	Check tool dashboard for error logs. Common cause: API rate limit exceeded. Reduce batch size or reschedule to off-peak.
False positive reported by sales rep	Review the flagged pair in tool. If legitimate non-duplicate, reject and note pattern. If pattern repeats, adjust matching threshold.
Duplicates re-entering after cleanup	Check for new data sources without prevention rules. Verify form submissions and imports have dedup checks active.
Opt-out preference lost after merge	Check survivorship rules — opt-out must be "keep most restrictive." If misconfigured, fix immediately and notify compliance. Attempt unmerge if available.
HubSpot-Salesforce sync creating duplicates	Verify HubSpot "auto-create company" setting is disabled. Check sync conflict resolution rules. Ensure dedup tool runs after sync, not before.
Manual review queue growing faster than team processes	Raise auto-merge confidence threshold to reduce queue volume. Or add reviewers. Or increase automation frequency.

For Dedicated engagements: Architect also receives the maintenance schedule (4a) and becomes responsible for executing it. SME walks Architect through each maintenance task in the internal handoff meeting.

---

4c. External Handoff (Delivery Team → Customer)

Purpose: Formal project completion with customer.

Final project meeting:

• Review what was delivered: tool configured, rules active, initial backlog cleared, automation running
• Walk through documentation package
• Walk through maintenance schedule in detail (monthly, quarterly, annual tasks)
• Demo unmerge/rollback one more time as a safety refresher
• Confirm nothing outstanding
• Answer final questions
• Make it explicit: "Project complete"
• For Single Project engagements: Hand over the maintenance schedule and walk through each task. Record a video walkthrough of the maintenance process.

Documentation package:

• Tool Configuration Runbook (all rules, thresholds, settings documented)
• Automation Schedule Document (what runs when, notification recipients)
• Manual Review SOP (merge/reject criteria, escalation triggers)
• Quick-Reference Guide (common tasks 1-pager)
• All training video walkthrough recordings (3 videos)
• FAQ document
• Definition Alignment Document (final version)
• Maintenance Schedule
• Support contact info

Output: Customer owns the deduplication tool and ongoing management. Project formally complete.

---

4d. Project Close

Purpose: Clean internal wrap-up + establish retention/expansion path.

Archive Checklist

• All project artifacts saved to proper location (audit report, configuration runbook, training materials)
• Handoff documentation complete
• Project status updated in tracking system
• Time/billing finalized

Internal Debrief (Optional but Recommended)

• What went well? (Tool selection speed, rule accuracy, training effectiveness)
• What would we do differently? (Better test data, earlier compliance involvement, different tool)
• Any learnings to feed back into SOPs? (New matching patterns, tool-specific quirks)

Retention / Expansion

Two paths based on engagement type:

Engagement Type	Path
Single Project	Upsell → Downsell → Retry
Multi-Project (Dedicated)	Schedule Refinement Check-In

Single Project Path:

1. Upsell: Managed Services (retainer -- ongoing dedup monitoring + rule tuning)
   | if no
2. Downsell: Another one-time project (CRM Data Enrichment, Lead Routing, Data Governance)
   | if yes
3. Retry retainer at end of next project cycle

Multi-Project (Dedicated) Path:

Schedule a refinement check-in at handoff:

"On [date ~quarter out], we'll review how the dedup tool is performing — duplicate rate trends, queue health, whether matching rules need adjustment."

Internal prep (2 weeks before check-in):

Step	What Happens
1. Get pinged	System reminder: refinement check-in in 2 weeks
2. Review metrics	Pull duplicate rate trend, queue backlog, automation run history
3. Decide ownership	Can Architect handle this check-in, or need SME?
4. Prep materials	If SME needed, brief them. If Architect, prep talking points.

At the refinement check-in:

• Review duplicate rate against target (<5% ongoing, <2% best-in-class)
• Check manual review queue health
• Identify any rule adjustments needed
• If minor: Architect handles threshold tweaks
• If major: Scope refinement project (new objects, new matching patterns, tool migration)

Output: Project archived. Future revenue path established. Ready for next engagement.

---

Deliverables & Assets Summary

Strategic Deliverables:

• Duplicate Audit Report (baseline rates, sources, patterns per object)
• Tool Comparison Matrix (evaluated options with recommendation)
• Definition Alignment Document (signed-off terms)
• Configuration Spec (matching rules, survivorship, thresholds, automation schedule)

Technical Deliverables:

• Configured deduplication tool with active matching rules for Contacts/Leads and Accounts/Companies
• Master record survivorship rules configured (field-level, opt-out preservation)
• Real-time duplicate prevention active on record creation, forms, and imports
• Automated merge workflow running on defined schedule
• Manual review queue with assignment rules and review criteria
• Monitoring dashboard (duplicate rate trending, merge volume, queue backlog)

Documentation Package:

• Tool Configuration Runbook
• Automation Schedule Document
• Manual Review SOP
• Quick-Reference Guide (common tasks 1-pager)
• Training video walkthrough recordings (3 videos)
• FAQ document
• Definition Alignment Document (final version)
• Maintenance Schedule

---

Appendix: Reference Guide

Phase Outputs & Gate Criteria

Phase	Output	Gate Criteria
Phase 1: Strategy	Tool selected, configuration spec signed off (rules, survivorship, thresholds)	RevOps and compliance stakeholders approved, tool procured
Phase 2: Engineering	Configured and tested dedup tool, initial backlog cleared	All rules tested, false positive rate <5%, customer approved preview merge
Phase 3: Enablement	Trained team with documentation, stabilized tool	All training delivered, 30-day hypercare complete, team operating independently
Phase 4: Handoff	Independent customer + archived project	Internal/external handoffs complete, maintenance plan in place, project closed

Project Profile

Project Profile	Strategy Weight	Engineering Weight	Enablement Weight	Notes
CRM Dedup Tool	20-30%	50-60%	15-20%	Engineering-heavy: tool config is the core work

Key Data Points

Organizations lose an estimated $12.9 million annually due to poor data quality [5]. In a typical B2B CRM, 10-30% of records are duplicates [6], and sales reps lose approximately 27% of their working hours — about 546 hours per year — dealing with bad or duplicate prospect data [4]. Best-in-class organizations maintain duplication rates below 2%, while anything above 5% signals a significant problem [6].

CRM-Specific Technical Notes

Salesforce Duplicate Management

Salesforce provides native Duplicate Rules and Matching Rules [2]:

• Up to 5 active Duplicate Rules per object
• Up to 5 active Matching Rules per object (3 matching rules per duplicate rule)
• Duplicate Rules can warn users, block creation, or allow with alert
• Duplicate Rules do NOT fire on API-created records, recycle bin restores, or manual merges
• Third-party dedup tools fill gaps: fuzzy matching, automation, cross-object, bulk operations

Start with warning-only mode to assess volume before enforcing blocks.

HubSpot Duplicate Management

HubSpot provides native deduplication [3]:

• Automatic dedup on email address for Contacts
• Duplicate Manager tool requires Professional or Enterprise subscription
• Bulk duplicate management requires Operations Hub Professional or Enterprise
• Native tool does not support automated scheduled merges
• Third-party tools (Insycle, Koalify, Dedupely) add automation, custom matching rules, and scheduling
• Disable "auto-create company from contact activity" if syncing with Salesforce

Cross-Platform Sync Considerations

When the client runs both HubSpot and Salesforce with a sync:

• Determine source of truth per object (usually Salesforce for Accounts, HubSpot for marketing Contacts)
• Run dedup AFTER sync completes, not before — otherwise sync re-creates merged records
• Align dedup rules across both platforms to prevent one system's merge from conflicting with the other
• Disable HubSpot auto-create company setting to prevent phantom Account duplicates

Troubleshooting Reference

Scenario	Symptoms	Root Cause	Resolution
Automated merge run failed	Error notification, no merges completed	API rate limit exceeded, CRM maintenance window, expired credentials	Check tool error logs. If rate limit: reduce batch size or reschedule off-peak. If credentials: re-authenticate.
False positives flagged by users	"These aren't the same person"	Matching threshold too low, fuzzy matching too aggressive	Review flagged pair. If pattern: raise threshold for that rule. If one-off: reject and move on.
Duplicates keep appearing after cleanup	Duplicate count rising despite automation	New data source without prevention rules, import process bypassing checks, sync re-creating records	Audit all data entry points. Configure prevention for any unprotected source. Check sync timing.
Opt-out/consent data lost after merge	Compliance alert, contacts receiving emails after opting out	Survivorship rule not preserving most restrictive opt-out	Fix survivorship rule immediately. Notify compliance. Attempt unmerge. Audit all recent merges for same issue.
Wrong master record selected	Important record data overwritten	Survivorship criteria not matching business intent	Review master selection logic. Unmerge if possible. Adjust survivorship rules. Re-test with similar scenarios.
Tool not detecting known duplicates	Manual review finds missed duplicates	Matching fields too narrow, fuzzy tolerance too strict	Expand matching criteria (add secondary fields). Lower fuzzy threshold. Consider cross-field matching.
HubSpot-Salesforce sync conflict	Records merged in one system reappear as duplicates in the other	Sync runs before or during dedup. Auto-create company setting still enabled.	Reschedule dedup to run after sync. Verify auto-create setting disabled. Align rules across platforms.
Manual review queue overwhelming team	Queue growing faster than team can review	Auto-merge threshold too conservative	Raise auto-merge confidence threshold. Add reviewers. Increase automation frequency.

---

References

[1] HubSpot - Deduplication of Records

[2] Salesforce Ben - Complete Guide to Salesforce Duplicate Rules

[3] HubSpot - Operations Hub Duplicate Management

[4] SalesForge - How Duplicate Data Impacts Sales Outreach

[5] Gartner via Databar - Duplicate Record Management in CRM: The Hidden Revenue Killer

[6] Landbase - CRM Match Rate Statistics: 25 Key Facts Every B2B Marketer Should Know

[7] Insycle - HubSpot Deduplication & Integration Tools

[8] Openprise - CRM Data Deduplication Software