HubSpot Data Hygiene

Data hygiene is the ongoing process of keeping your HubSpot database clean, accurate, and useful. Duplicate contacts, stale deals, missing fields, and incorrect properties degrade every system that depends on CRM data: scoring, routing, reporting, outreach, and forecasting. Clean data is the foundation of every GTM operation.

The principle: data hygiene is not a one-time project. It's a recurring process. Data degrades at 2-3% per month through job changes, company merges, email bounces, and human error. A database that's clean today is 25-35% degraded in a year without ongoing maintenance.

The Data Hygiene Framework

Four dimensions of data quality

The Hygiene Audit

Step 1: Assess current state

Step 2: Quantify the problem

Database health scorecard:
  Total contacts: 50,000
  Duplicates: 4,200 (8.4%)
  Invalid emails (bounced/unverifiable): 6,500 (13%)
  Missing company association: 8,000 (16%)
  Missing job title: 12,000 (24%)
  No activity in 12+ months: 18,000 (36%)
  
  Health score: 100 - (8.4 + 13 + 16 + 24 + 36)/5 = 80.5%
  Target: > 90%

Deduplication

Finding duplicates

Deduplication rules

• Merge, don't delete. When you find duplicates, merge them. Deleting loses activity history, form submissions, and deal associations. Merging preserves everything on the surviving record
• Keep the record with the most data. When merging, the record with more populated properties, more activities, and more recent engagement should be the surviving record
• Automate ongoing dedupe. Run HubSpot's dedupe tool monthly. Set up a workflow to flag new contacts that match an existing email or company+name pattern
• Fix the source. Duplicates enter through forms without email matching, imports without dedup checks, and integrations without proper matching. Fix the entry point, not just the symptom

Email Hygiene

Email validation process

1. Export all contacts with email addresses
2. Run through an email verification tool
   (NeverBounce, ZeroBounce, or equivalent)
3. Results:
   - Valid: keep as-is
   - Invalid: mark as "Email Invalid" in HubSpot
   - Risky: re-verify in 30 days
   - Catch-all: keep but monitor bounce rate
4. Suppress invalid emails from all sending
5. Re-verify the full database every 6 months

Email hygiene rules

• Verify before sending. Never send a cold outbound campaign without verifying the list first. A 10%+ bounce rate damages your sending domain reputation
• Remove hard bounces immediately. When HubSpot logs a hard bounce, mark the email as invalid and suppress from all sequences. Don't wait for the monthly cleanup
• Re-verify every 6 months. Email addresses go stale. People change jobs, companies change domains. A valid email 6 months ago may bounce today
• Track bounce rate as a hygiene metric. Bounce rate > 5% on any campaign means your email data needs attention. Target < 2%

Property Cleanup

Common property issues

Property cleanup process

Quarterly property audit:
1. Export all custom properties with fill rates
2. Flag properties with < 10% fill rate
3. For each flagged property:
   - Is it used in any report? → Keep if yes
   - Is it used in any workflow? → Keep if yes
   - Is it required on any form? → Keep if yes
   - None of the above → Archive
4. Review "Other" fill rates on dropdown properties
   - If > 15% are "Other", add missing options
5. Check for duplicate-purpose properties
   - Merge or standardize

Stale Record Management

Contact staleness

Deal staleness

Staleness rules

• Auto-alert on stale deals. Workflow: if deal has no activity for 14 days and is in an active stage, notify the deal owner. Don't wait for the monthly pipeline review
• Auto-close zombie deals. If a deal has no activity for 60 days and the owner doesn't update after 2 alerts, auto-close as Lost with reason "Stale - No Activity." This keeps pipeline real
• Don't delete stale contacts. Suppress them from sending. Mark them as "Cold" or "Inactive." They may re-engage. Deleting loses all history

Ongoing Hygiene Cadence

Measurement

Pre-Cleanup Checklist

• [ ] Database health scorecard calculated (current state assessment)
• [ ] Duplicate detection run on contacts and companies
• [ ] Email verification run on the full database
• [ ] Property fill rates exported and reviewed
• [ ] Stale deals identified (no activity 30+ days)
• [ ] Stale contacts identified (no activity 12+ months)
• [ ] "Other" dropdown values reviewed for top 10 dropdown properties
• [ ] Unused properties identified (< 5% fill rate, no report/workflow usage)
• [ ] Hygiene cadence assigned (who does what, how often)
• [ ] Automated workflows configured (duplicate alerts, stale deal alerts, bounce suppression)

Anti-Pattern Check

• One-time cleanup, no ongoing process. You spend 40 hours cleaning the database. It's perfect. 6 months later, it's 20% degraded. Data hygiene is recurring. Assign a monthly cadence and an owner
• Deleting duplicates instead of merging. You find 3,000 duplicates and delete 1,500 contacts. You lost all their activity history, form submissions, and deal associations. Always merge. Never delete
• Ignoring bounced emails. Hard bounces pile up. Sending reputation degrades. Deliverability drops from 95% to 80%. All because nobody suppressed the bounced addresses. Remove hard bounces immediately
• No validation on data entry. Forms accept any text in any field. Integrations push unvalidated data. Imports have no dedup check. Clean data starts at the point of entry. Validate before it hits the CRM
• 200 custom properties, half unused. Every initiative added properties. Nobody archived them. Forms have 30 optional fields. Contact records are unreadable. Audit quarterly. Archive anything unused
• Stale deals inflating pipeline. 40% of open deals have no activity in 45 days. Pipeline report says $3M. Real pipeline is $1.8M. Clean stale deals bi-weekly. Auto-close after 60 days of inactivity
• Data hygiene is "someone else's job." Marketing blames sales for bad data. Sales blames marketing. Nobody cleans anything. Assign a specific owner (usually RevOps) with a specific cadence and specific metrics