TL;DR
HubSpot’s AI-powered Manage Duplicates tool (Starter and above) surfaces duplicate contacts using email, name, phone, company, and IP signals. Data Hub Professional and Enterprise portals can bulk merge up to 50 contacts at a time. Enterprise HubSpot portals typically carry 3 to 8 percent duplicate rates that break reports, workflows, and Breeze AI outputs. This guide covers how to find them, merge them, and prevent them from returning.
Why HubSpot duplicate contacts matter
Duplicate contacts are the most damaging data quality issue mid-market HubSpot portals face. They distort marketing attribution, inflate contact counts (which affects your billing tier), corrupt lifecycle stage reporting, and now cause AI agents to reason across conflicting versions of the same person.
That last consequence is new and it is the one changing the urgency. An agent reading two records for the same contact does not flag a conflict, it picks one and answers confidently. We covered the wider pattern in why AI agents fail in RevOps: the model is rarely the problem, the record underneath it is. If you are planning a Breeze AI deployment, duplicates are a blocker rather than a housekeeping task.
HubSpot’s native deduplication has improved, but it still needs an active management strategy. Deduplication is one part of the broader data quality work covered in our CRM optimization guide.
Why HubSpot duplicate contacts happen
Duplicates rarely come from one source. They accumulate from several entry points, and a cleanup only stays clean if you close all of them.
1. Form submissions from different browsers or devices
When a prospect submits a form, HubSpot checks their browser cookie against existing contacts. If the same person submits from a different browser, incognito mode, or another device, HubSpot creates a new record instead of updating the existing one.
2. CSV imports without deduplication rules
Teams importing purchased lists, event attendees, or webinar registrants routinely create duplicates. Unless the option to avoid creating duplicate contacts is enabled, the import creates a new contact for every row.
3. Salesforce or other CRM sync creating parallel records
A misconfigured sync mapping creates parallel contacts on both sides. This is the most common cause in enterprise portals running two CRMs, and the mapping detail matters more than the tooling. Our HubSpot Salesforce integration admin guide walks the object mapping and sync rules, and our multi platform RevOps comparison covers when running both systems is worth the overhead.
4. Job changes and email address updates
A contact leaves their company, changes email, and re-engages under a new address. HubSpot has no way to know these are the same person, so a new contact is created while the old one goes stale.
5. Chat conversations creating orphan contacts
Chatbots and live chat create new contacts when visitors are not cookie-matched to an existing record. High traffic sites accumulate hundreds of orphan chat contacts monthly.
6. Manual entry by sales reps
Reps adding contacts without searching first is the oldest source of duplicates and still one of the biggest. Portals with 20 or more reps accumulate manually created duplicates every week.
How to find duplicate contacts in HubSpot
HubSpot’s Manage Duplicates tool scans your database daily and surfaces likely duplicates using email address, first and last name, phone number, company name, and location signals, generating a confidence score for each suggested pair.
1. Go to CRM and click Contacts.
2. Click the Actions dropdown above the contact table.
3. Select Manage duplicates.
4. Review the suggested pairs, ranked by confidence (High, Medium, Low).
5. Filter by confidence to work through high confidence matches first.
6. Click Review on each pair to inspect both records side by side.
7. Export the list first if you want to audit findings before merging anything.
The tool displays a capped number of duplicate pairs depending on your tier, rising with Data Hub Professional and Enterprise. If your portal is at the cap, that is a signal the problem is structural rather than incidental. A broader portal review is covered in our HubSpot audit checklist.
How to merge duplicate contacts in HubSpot
There are two paths: a one-by-one merge from any contact record, and the bulk flow described below. To merge individual contacts:
1. Open the contact you want to keep as the primary record.
2. Click Actions at the top left of the contact page.
3. Select Merge.
4. Click Select contact and search for the duplicate.
5. Choose the correct duplicate from the results.
6. Review property values side by side and choose which to keep.
7. Click Merge.
8. Verify associated companies, deals, and tickets carried over.
| Merges are permanent. HubSpot does not support un-merging. There is no undo, no 30 day window, and no support ticket that reverses it. Review both records fully before you click. |
How to bulk merge HubSpot duplicates
For portals with hundreds or thousands of duplicates, merging one pair at a time does not scale. There are three approaches, all requiring paid plans.
Option 1: Native bulk merge (Data Hub Professional or Enterprise)
1. Go to CRM > Contacts.
2. Click Actions and select Manage duplicates.
3. Use the checkboxes to select up to 50 duplicate contacts.
4. Click Review on the selected pairs.
5. Choose which record stays as primary.
6. Click Merge All.
Option 2: Data Hub programmable automation
Data Hub Professional and Enterprise include programmable automation that can detect duplicates against custom logic, such as matching email plus phone, or domain plus name, and merge them through a workflow. This is the most powerful native option for high volume portals, and the one most Enterprise customers are paying for without using.
Option 3: Third party dedup apps
Tools such as Insycle and Dedupely offer probabilistic matching, custom merge rules, and dedup automation beyond the 50 record ceiling. They sit outside the HubSpot UI, which adds a learning curve, but they handle scenarios native HubSpot cannot.
What data is kept or lost during a merge
This is the section most guides skip, and the one that causes the most regret. Understanding what survives a merge is what separates a cleanup from an accidental deletion of your attribution history.
The takeaway: choose the primary contact deliberately. The primary record’s original source, first touch attribution, and property values dominate the merged contact. If the secondary record carries the more valuable attribution history, promote it to primary before merging. Teams that skip this step spend the next quarter arguing about channel performance that no longer reconciles, which is the same failure described in our piece on how attribution reporting impacts marketing ROI.
How to prevent HubSpot duplicates going forward
Merging is remedial. Prevention is where the return sits. Each item below closes one of the six causes above.
Block duplicate creation on every import
The single highest impact setting. Every CSV import should have duplicate creation disabled. Put it in the import SOP so it survives staff turnover, not just in someone’s head.
Standardize forms on one primary email property
Every form should write to the same email property. Separate “Work email” and “Business email” custom properties fragment identity across the database and guarantee duplicates.
Audit sync mappings quarterly
Sync errors accumulate silently. Review the error queue at least quarterly and repair mappings before they generate more duplicates. This is standard scope in a properly run HubSpot implementation.
Set chatflow contact creation rules
Require an email address before a chatbot creates a contact. Anonymous chat leads are rarely worth the duplicate volume they generate.
Train reps to search before creating
Add a mandatory search step to rep onboarding, and reinforce it with monthly reporting on manually created duplicates by rep. Visibility changes behaviour faster than training does.
Run a monthly dedup review
Assign one named owner. Whether that sits with RevOps, marketing ops, or a CRM admin depends on how your team is structured, which we unpack in RevOps vs sales ops vs marketing ops. High volume B2C portals should run it weekly. Prevention plus regular review holds duplicate rates under 2 percent.
When to escalate a HubSpot dedup project
Bring in outside RevOps support when any of these are true:
• Your duplicate rate is above 5 percent of total contacts
• Duplicates are causing forecast drift or attribution disputes
• You are preparing to deploy Breeze AI or another agent capability
• You are planning a Salesforce migration or acquisition due diligence
• Integrations are creating duplicates faster than you can merge them
• You hold Data Hub Enterprise but are not using programmable dedup automation
Deduplication is usually a symptom rather than the disease. If duplicates keep returning after cleanup, the underlying issue is architectural, and that is the scope of our CRM implementation and enhancement work. Comparable engagements are documented in our case study portal.
Audit your HubSpot duplicatesIf your portal holds more than 500 contacts and you have not run a dedup review in six months, you almost certainly have architectural debt affecting your reporting and blocking any future AI deployment. The Mountainise RevOps AI audit scans your portal for duplicate rate, association integrity, workflow conflicts, and 12 other CRM architecture issues in minutes. Free at the diagnostic level, and no consultant call required. Connect your portal, read the report, decide what to fix first. Prefer to talk it through? Get in touch. |
Frequently Asked Questions
Go to CRM > Contacts, click Actions, and select Manage Duplicates. HubSpot surfaces likely matches ranked by confidence (High, Medium, Low) from a daily scan. Filter by confidence and work through high confidence matches first.
No, not by default. HubSpot scans daily and surfaces suggestions for review. Data Hub Professional and Enterprise include programmable automation that can auto-merge against custom rules, but you have to build it.
Yes. Data Hub Professional and Enterprise allow bulk merging up to 50 contacts at a time through Manage Duplicates. Programmable automation handles higher volumes, and third party apps such as Insycle and Dedupely support larger bulk operations with custom matching rules.
No. HubSpot merges are permanent and there is no un-merge function. This is why choosing the primary record matters so much: the secondary contact’s original source and first touch attribution are deleted at the moment you merge, and no support request will bring them back.
The primary contact’s original source and first touch attribution are kept, and the secondary contact’s equivalents are lost. Timeline activity and associated companies, deals, and tickets are combined. Property values default to the primary record unless you override them during the merge.
HubSpot matches form submissions to existing contacts by browser cookie, so a different browser or device produces a new record. Enforce one primary email property across every form, avoid additional email custom properties, and disable duplicate creation on all CSV imports.
Six things: form submissions from different browsers or devices, CSV imports without dedup rules, misconfigured CRM sync mappings, contact job changes and email updates, chat conversations creating orphan contacts, and manual entry by reps who did not search first.
Monthly is sufficient for most teams. High volume B2C brands and organizations running heavy paid acquisition should scan weekly to keep contact counts and marketing sends accurate.