Loading tutorials…
Loading tutorials…
Pipedrive data quality degrades silently. By month 6, you have 800 duplicate contacts, 200 deals nobody owns, and a forecast nobody trusts. Here is the cleanup playbook — and the hygiene cadence that prevents the next round.
Who this is forSales ops leads and founders looking at a Pipedrive account they suspect is full of junk. If duplicate-management is on your weekly to-do list, or your forecast misses by 30%+ every quarter, this tutorial is the diagnostic.
What you'll need
Step 1
Tools and integrations → Duplicates. Pipedrive auto-detects likely duplicates by email (Person) and domain (Organization). The list usually shocks people.
Open Tools and integrations (left sidebar) → Duplicates.
Pipedrive surfaces likely duplicates for People (matched on email) and Organizations (matched on root domain).
Review the suggested merges in batches of 20-50. For each, click into both records — confirm they are actually the same person/org, then merge.
Merging preserves all associated deals, activities, notes, and emails. The "primary" record (the one you keep) inherits everything from the secondary.
For names without matching emails (e.g., "John Smith" appearing twice with different emails), Pipedrive does not auto-detect — these require manual search and merge.
Step 2
Cleanup without source-fix means duplicates return. Find which workflow or integration is creating them.
Common source 1: web forms that create new Person records without checking for email match first. Audit your LeadBooster web forms and any third-party form tools — most have a "match existing contact by email" toggle that needs to be On.
Common source 2: Zapier or Make.com flows that "create person if not found" without proper search logic. Audit Zapier zaps that write to Pipedrive — confirm they search by email before create.
Common source 3: reps manually creating Persons instead of searching first. Train this in the team SOP. For SDRs especially, restrict person-create permissions to reduce sprawl.
Common source 4: email sync creating contacts from every sender (the "auto-create from email" setting). Personal preferences → Email sync → turn this Off across the team.
Common source 5: CSV imports that lack proper dedup checking. Always dedup the CSV in Excel/Sheets before importing. Pipedrive imports with the "Update existing items if matched" toggle enabled.
Step 3
Filter deals by Last activity older than 60-90 days. For each, either revive with a concrete next step or mark Lost.
Deals view → filter → "Last activity date is more than [60] days ago AND Status is Open." Sort by Last activity ascending.
For each stale deal: ask the owner directly — is this real? If yes, what is the next step and when? Schedule the next activity right then.
If no, mark Lost with a specific reason. Use the Lost reason field (Company settings → Deal settings → Lost reasons). Common: 'Stalled — no buyer urgency,' 'No response,' 'Lost to competitor,' 'No budget.'
Forecasting accuracy degrades 10-15% per quarter with stale deals piling up in the weighted pipeline. The number 'expected to close' includes 6-month-old proposals nobody is working.
Run this audit monthly. Make it a calendar recurrence. Stale-deal review is the single highest-leverage hygiene practice in any CRM.
Step 4
Records owned by deactivated users (former employees) become orphans. Reassign before deactivating users; clean up the orphans that already exist.
Deals view → filter → "Owner is [deactivated user name]." Same for Persons and Organizations views.
Bulk reassign: select rows → Edit → Owner → pick new owner. Pipedrive allows up to 5,000 records per bulk edit.
When deactivating a user (Company settings → Manage users → deactivate), Pipedrive prompts to reassign their records. Always do this BEFORE deactivation, never after.
For records owned by old users that you forgot to reassign: filter by deactivated owner, bulk-reassign to a current owner or to a generic "Unassigned" admin account.
Document a SOP: "Before deactivating a user, reassign all their open deals, all their persons, all their organizations." This prevents orphan accumulation.
Step 5
Free-text fields drift. "Industry" might contain "SaaS," "Software," "software," "B2B SaaS," "Tech." Standardize.
Deals/Persons/Organizations view → group by the field in question. Pipedrive shows the distinct values and the count of records per value.
For Text fields that should be categorical, look for variations: 'United States,' 'USA,' 'US,' 'America,' 'U.S.' — all the same country.
Bulk update: filter to the variant value → select all → Edit → set to the canonical value. Or convert the field type from Text to Single option (this preserves data but only if the type-change is safe — back up first).
Most reporting bugs in Pipedrive trace to dirty field values. 'Deals by industry' shows 14 industries when you only have 6 — the 8 extras are spelling variants.
After cleanup, prevent re-introduction: convert categorical Text fields to Single option (locked dropdown). Single option prevents future typos.
Step 6
Cleanup once = back to mess in 90 days. The cadence is what prevents re-accumulation.
Monthly recurring calendar block (1 hour). Owner: sales-ops lead or a delegated rep.
Checklist: (1) Run Duplicates view, merge new dupes. (2) Filter stale deals (no activity in 60+ days), force disposition. (3) Filter orphan records, reassign. (4) Spot-check dirty fields on top-priority Text/categorical fields. (5) Run Insights → Activity log: any new junk-data sources?
Quarterly: full data audit. Walk every custom field, check fill rates, prune fields under 25% fill. Audit Workflow Automations for stale workflows. Review Lost reason distribution.
Document the cadence in a shared doc. The discipline is what makes the data trustworthy long-term.
Track 3 hygiene metrics quarter-over-quarter: % of deals with no activity in 30+ days (target <10%), # duplicate persons created per month (target <20), # orphan records (target 0). When these drift up, the cadence is slipping.
Common mistakes
Never running the duplicate audit
What goes wrong: By month 12, 15-25% of Person records are duplicates. Search returns multiple matches for the same human. Email sync attaches conversations to whichever duplicate it finds first. Activity history is scattered across 3 records of the same person.
How to avoid: Tools and integrations → Duplicates → monthly cleanup. Plus: prevent at source — fix web-form dedup logic, audit Zapier zaps, turn off email-sync auto-create.
Letting stale deals sit forever in the pipeline
What goes wrong: Loses 10-15% of forecast accuracy with stale deals padding the weighted pipeline. Leadership keeps predicting revenue that does not materialize. Eventually the forecast becomes a number nobody believes.
How to avoid: Monthly stale-deal audit. Force disposition: revive with a concrete next step, or mark Lost with reason. Make it a calendar recurrence.
Deactivating users without reassigning their records
What goes wrong: Former employee left 60 days ago. Their 40 open deals and 200 contacts are still owned by their deactivated account. No reports include them (filter on 'active users' excludes deactivated). Pipeline shows artificially low.
How to avoid: SOP: before any user deactivation, bulk-reassign records. After deactivation, filter by deactivated user and reassign anything missed.
Categorical data in Text fields
What goes wrong: Industry, Country, Tier, Segment all stored as Text. Reps spell them differently. Reports show 14 industries when 6 exist. Marketing-sales handoff breaks because both sides use different category labels.
How to avoid: Convert categorical Text fields to Single option (Company settings → Data fields → field type). Back up data first. After conversion, locked dropdown prevents future typos.
Web forms creating new Persons without email match check
What goes wrong: Every inbound form submission creates a new Person — even when that person already exists. Within 60 days, 30% of inbound leads are duplicates of existing contacts. The 'is this a new lead?' question takes manual research.
How to avoid: Audit every form (LeadBooster, web forms, third-party tools like Typeform/Jotform). Enable "Match existing contact by email" before creating new. Zapier zaps: add a Find Contact step before Create Contact.
No documented hygiene cadence
What goes wrong: One person does cleanup heroically every quarter. They leave the company or get busy. Cleanup stops. Within 6 months the mess is back. The institutional knowledge of what 'clean' looks like is lost.
How to avoid: Document the monthly cadence checklist in a shared SOP. Assign an owner. Calendar-recurring block. Track 3 hygiene metrics quarter-over-quarter — drift is your early warning.
Recap
Done — what's next
How to set up Pipedrive custom fields that hold up at scale
Read the next tutorial
Hand it off
Data quality is the invisible CRM tax. Most owners do not realize how much forecast accuracy and SDR time it eats until they fix it. A specialist will run the cleanup + set up the monthly cadence + document the SOP — typically a $300-600 engagement that pays back the first quarter.
See specialist rates
Partially. Pipedrive auto-detects duplicates by exact email match (People) and root domain match (Organizations) and surfaces them in Tools → Duplicates. You still confirm and click Merge. Pipedrive does not auto-merge without confirmation — which is appropriate; auto-merging is dangerous.
All activities, notes, emails, deals, and files from the secondary record transfer to the primary. Conflicts on individual field values default to the primary's value (you can override per-field during merge). The secondary record is archived but not deleted — you can find it in the Audit log for 30 days.
Deals view → filter to the target set (e.g., 'Last activity > 180 days ago AND Status = Open') → select rows → Edit → Status = Lost + Lost reason = 'Stalled — no response.' Up to 5,000 deals per bulk edit. Always export the list first as a backup.
Pipedrive dedups Organizations on root domain, not on name. 'Acme Inc' (acme.com) and 'Acme, Inc.' (acme.com) auto-dedup. 'Acme Inc' (no domain) and 'Acme Inc' (no domain) do NOT auto-dedup because Pipedrive has no key. Always require Domain or Website on new organization creation to prevent this.
Monthly cadence: duplicate audit, stale-deal review, orphan reassignment, spot-check categorical fields. Quarterly: full audit including custom-field fill rates, automation review, Lost-reason distribution. Without this cadence, data quality degrades 5-15% per quarter on average.
Pipedrive
Custom fields are how Pipedrive bends to your business. They are also how owners create chaos in 60 days — 47 fields, half empty, three that say the same thing. Here is the discipline that prevents the cleanup project.
Pipedrive
The Pipeline view is the single most-referenced screen in Pipedrive. Set it up wrong, the forecast number is fiction and reps stop trusting it. Set it up right, and the Monday-morning pipeline review takes 20 minutes instead of an hour.
Pipedrive
Workflow Automation is the single biggest force multiplier in Pipedrive Professional+ — and the easiest feature to misuse. One bad automation can email 4,000 prospects at 3am. Here is the discipline that scales.
HubSpot CRM
Contacts, Companies, and Deals are the three records every HubSpot motion runs on. Get the relationship between them right and your reports are clean for years. Get it wrong and you spend Saturdays untangling duplicate companies and orphaned deals. Here is the discipline.
Pipedrive
DIY Pipedrive is a great idea — until it isn't. This is the honest framework: when the cost of self-managing exceeds the cost of hiring a specialist, and how to tell which side you are on.