Deduptio

Attio deduplication guides

Practical, Attio-specific answers to the duplicate questions that come up in real workspaces. Every page states what Attio does natively before it says anything about Deduptio, because in a fair number of cases the native answer is the whole answer.

Start here

Attio deduplication: the complete guide
The pillar: why duplicates appear, what Attio handles natively, how to design match rules, choose survivors, merge safely and verify the result.

One question, one page

The task-shaped guides. Each answers a single question — how to merge, how to find, how to undo — with the limits stated up front rather than discovered halfway through.

How to merge records in Attio
The native merge, step by step — including the three limits nobody mentions until you hit them: two at a time, people and companies only, and no undo.
How to bulk merge duplicates in Attio
Attio has no bulk merge in the UI and its merge API takes two records per call. How to clear a backlog of hundreds of groups without merging them by hand.
Find duplicate companies in Attio
Domain matching catches the easy half. Rebrands, acquisitions, agencies, holding companies and free-email founders need a rule that survives all five.
Find duplicate people in Attio
Email is the strongest signal a person record has and it changes every time they change jobs. How to match people when the identifier moved.
Duplicate deals in Attio
Attio cannot merge deals at all. What that means for a pipeline with the same opportunity entered twice, and the two ways out that do not lose forecast history.
Fuzzy matching company names
Why 'Acme Ltd' and 'ACME Limited' are the same company and 'Acme Health' is not — normalization, similarity thresholds, and the pairs that break every threshold.
Duplicate list entries in Attio
A record sitting twice in the same list is a different bug from a duplicate record — and merging duplicates is what usually creates it.
How to undo a merge in Attio
Attio's own answer is that you cannot: the losing record is deleted and unrecoverable. What is actually recoverable, by whom, and for how long.
Attio duplicate detection with the API
Assert, search and merge — what the REST API gives you for duplicates, what it costs in calls, and the four things you still have to build yourself.

Imports and migrations

Migrations are the largest single source of duplicates in a young Attio workspace, and almost all of it is preventable with the right load order and one setting turned off before you start.

Duplicates from a CSV import into Attio
Map a unique attribute and Attio updates instead of duplicating. Miss it and one import doubles an object — how to prevent it, and how to clean it up after.
Duplicates after a HubSpot to Attio migration
Email sync creating records mid-import is the number-one cause. The order to load objects in, and the cleanup pass to run once the data has landed.
Duplicates after a Pipedrive to Attio migration
Pipedrive organizations have no required domain and its people allow multiple emails — two structural mismatches that turn into duplicates on arrival.
Duplicates after a Salesforce to Attio migration
Leads and Contacts both become Attio people, Accounts become companies, and years of Salesforce duplicates arrive with them. The de-dupe order that works.

Also useful

See what is duplicated before changing anything

Connect your Attio workspace, write one match rule, and run a scan. Scans are read-only — you see every group and the reason it matched before anything is merged.

Start a free Attio duplicate scan