How to deduplicate HubSpot and Salesforce while they sync

Short answer: yes, you can merge duplicates while HubSpot and Salesforce sync, but each merge has to happen in both CRMs. Done by hand, that means merging in Salesforce first, keeping the record that syncs with HubSpot, waiting for the sync and then merging in HubSpot. With Dedupely, which connects to both CRMs, the two merges happen in one action.

This post covers why one-sided merges break the sync, the manual method step by step, what it takes to keep it running, and how Dedupely merges both sides at once.

What happens when you merge in only one CRM

HubSpot and Salesforce each keep their own records, and the sync links them in pairs. A merge only changes the CRM it's made in.

So a one-sided merge leaves a duplicate on the other side, or a record that has quietly stopped syncing. Either way, the duplicate is still there.

Why merged records come back

Salesforce allows two leads or contacts with the same email. HubSpot doesn't, and it syncs with whichever matching Salesforce record was updated most recently. If you merge in HubSpot but leave both Salesforce records in place, the sync keeps overwriting the HubSpot contact from two sources, and a later update can create the duplicate again.

The fix is the same in every case: the merge has to cover both CRMs, with the surviving records still linked to each other.

The manual method: Salesforce first, then HubSpot

This is the usual way to merge synced duplicates without a tool that's connected to both CRMs. It follows HubSpot's own advice: when you merge in Salesforce, keep the record that's currently syncing with HubSpot as the primary.

1. Stop merges from deleting HubSpot contacts. In HubSpot's Salesforce integration settings, set what happens when a Salesforce contact or lead is deleted to "Do nothing". Otherwise, merging in Salesforce deletes the HubSpot contacts linked to the merged-away records.
2. Find the record each side syncs with. On a HubSpot contact, the Salesforce Lead ID or Salesforce Contact ID property shows which Salesforce record it's linked to.
3. Add a field to mark the record you keep. Many teams create a checkbox field in Salesforce and a matching property in HubSpot, then map it so Salesforce's value always wins. It tells the HubSpot side which record survived the Salesforce merge. You need one per object you merge.
4. Merge in Salesforce. Keep the record that syncs with HubSpot as the primary, and tick the marker field on it.
5. Let the sync run. Check that the marker field has reached HubSpot before you go on.
6. Merge in HubSpot. Keep the marked contact or company as the primary.
7. Clean up what's left. With deletion set to "Do nothing", HubSpot contacts whose Salesforce record was merged away stay in HubSpot, unlinked. Review and merge or delete them. If a contact stopped syncing, use Resync on its Salesforce Sync card.

What the manual method takes

It works, and for a one-time cleanup of a few sets it's often enough. Over time, it adds up:

• Setup in both CRMs: a settings change, a field per object in each CRM and a field mapping.
• Two merges for every set, with a wait for the sync in between.
• A cleanup step for the records left unlinked in HubSpot.
• The same steps every time new duplicates come in.
• Companies and accounts are separate: HubSpot companies don't merge their Salesforce accounts, so they need the same process.

Which CRM to merge first

With the manual method, Salesforce first. When the Salesforce side is merged first, with the synced record kept as primary, each HubSpot contact still has the Salesforce record it syncs with, and the HubSpot merge only has to keep the right contact. If you merge in HubSpot first, the duplicate Salesforce records stay in place and can keep updating or bringing back the record you merged away.

When both CRMs are merged in one action, there's no order to manage.

Can you undo a merge?

Merges made in Dedupely in one CRM can be undone within 30 days from Merge History. The exceptions are merges made in HubSpot and Salesforce at once, and company merges in a HubSpot portal synced with Salesforce, so test those on a small batch first.

In the CRMs themselves:

• In HubSpot, a merge made in HubSpot is permanent, and by default the merged record gets a new Record ID.
• In Salesforce, the merged-away record goes to the Recycle Bin, which keeps deleted items for 15 days. Restoring it doesn't change the master record.

Before any merge in a synced setup, export the records you'll merge with their values and each record's ID in both CRMs.

What to settle before you merge

• Which value wins in each field when the two records disagree, like owner, lifecycle stage or phone.
• Which records to start with: exact email or domain matches first, fuzzier matches later.
• Where the original values are kept, so you can restore a field if a merge picks the wrong one.
• Permissions in both CRMs: the user that merges needs merge rights in HubSpot and in Salesforce.

One merge, two CRMs

Synced duplicates have to be merged on both sides, with the surviving records still linked. By hand, that means merging in Salesforce first, then HubSpot, with a marker field, a settings change and a cleanup step. Dedupely does both merges in one action, keeps the synced records as primary and logs every merge.

Frequently asked questions