Migrating into Intercom without duplicate contacts

Where duplicates come from

• Matching on the wrong thing. Two people can share a name, and one person can have different addresses in different systems. Identity has to come from stable identifiers.
• Records written at the same time. If two copies of the same person are written together, both look for a match, find none, and both create a contact.
• Contacts Intercom is hiding. In our testing, archived contacts did not appear in search but still blocked a new contact with the same details, and a deleted contact’s details stayed blocked for a while.

Tags: check the spelling

Intercom’s tag endpoint both creates and updates tags. In our testing, creating a tag whose name matched an existing tag in different capitals renamed the existing tag across the workspace. Read the workspace’s tags first and reuse each under its existing spelling.

Before you migrate

1. 1.Decide which identifier is the source of truth for a person: your own user ID as the Intercom external ID, or email.
2. 2.Look for archived contacts in the destination that may match migrated people.
3. 3.Avoid deleting contacts in the destination just before a migration.
4. 4.List the tags already in the destination, and how they are spelled.
5. 5.Run a sample and count contacts on both sides before the full run.

DataGrine migrations into Intercom start with a written field mapping and a sample run signed off before the full run, run self-serve on the platform or fully managed by the migration team, with 24/7 support either way.