Deduptio vs Attio's native merge

Last checked September 1, 2026 · Deduptio is our product. Every claim about another tool comes from its public site or documentation and is linked in place — tell us if something is out of date.

Attio has a merge built in. It is free, it is good, and Deduptio calls it — every merge Deduptio performs goes through Attio's own record merge, which is why emails, calls, notes, tasks and comments end up on the surviving record exactly as they would if you had merged by hand.

So this is not really a comparison of two ways to merge. It is a comparison of the merge on its own against the merge plus the four things Attio deliberately leaves out: finding duplicates that share no exact identifier, handling groups larger than two, doing hundreds in one pass, and undoing a mistake.

The short version: if you can name your duplicates, merge them in Attio and close this page. If you cannot — because nobody knows how many there are, because they sit on deals or a custom object, or because the number is in the hundreds — the native merge is not the constraint you are up against.

Side by side

What the native merge actually does

Open a company or person record, choose Merge records, search for the duplicate, and confirm. The two records become one. List entries, notes, tasks and comments from both are combined, and for people, email threads come along too. The record on the right is the one that persists; the one on the left is deleted. You can swap them before confirming, and you need write access to the object to do any of it.

For an obvious pair you have just noticed, nothing beats this. It takes ten seconds, it costs nothing, and the outcome is the canonical one — every other tool in this category, Deduptio included, is trying to reproduce it.

Three limits, and when each one starts to hurt

1. Two records at a time

A merge combines exactly two records. Duplicate groups are not always pairs: an imported contact, a form submission and a record created by email sync make three, and clearing that group is two merges. Attio's REST API has a merge endpoint with the same two-record shape, so scripting it removes the clicking but not the pairwise structure — and you are then responsible for ordering the chain so that a record consumed by one merge is not passed to the next. Deduptio merges a group by walking exactly that chain, keeping per-hop progress so an interrupted run resumes instead of re-merging.

2. People and companies only

Attio's merge is available for company and person records. It is not yet possible to merge deals, workspaces, users or custom objects. If your duplicates are two versions of the same opportunity, or two rows on a custom Properties or Applications object, the native merge has no answer at all — see duplicate deals in Attio for what the options actually are.

3. No undo

This is the one that changes behaviour. Attio's documentation is explicit: once a record is deleted it cannot be recovered by you or your team, and records previously merged into it are not restored either. So a native merge is a one-way door, and the rational response to a one-way door is to go slowly — which is why so many Attio workspaces have a duplicate backlog that everyone knows about and nobody clears.

Deduptio snapshots both records before it calls the merge. That is the whole trick: for 24 hours the pre-merge records can be re-created with their field values and associations. It is not time travel — the restored records are new records with new IDs, and the notes, tasks, list memberships, emails and calls that Attio moved onto the survivor are not re-created. Rollback spells out exactly what comes back and what does not, because a safety net you have misunderstood is worse than none.

The part that is not about merging at all

Attio prompts you to merge when it spots a duplicate by exact email or exact domain. That catches the easy half. It does not catch a person whose email changed with their job, a company that operates on two domains, a name typed two ways, a record imported without an email at all, or anything on an object Attio does not detect duplicates for.

Detection and resolution are separate jobs, and the native merge only does the second one. Most workspaces do not have a merging problem; they have a knowing what to merge problem. The Attio deduplication guide covers how to design rules that reflect identity rather than similarity.

Which should you use?

• Native merge when you have found a duplicate pair during normal work, it is a person or a company, and you are confident about which record should survive. It is free and it is right there.
• Deduptio when you do not know how many duplicates you have, when they are on deals or custom objects, when the groups are bigger than two, when there are more than a few dozen, or when you want the merge to be reversible while you learn what your data actually looks like.

The honest test: try to merge ten of them by hand. If that clears the problem, you are done and you have spent nothing. If it makes barely a dent, the backlog is a scanning problem and no amount of clicking will finish it.

Related

• How to merge records in Attio — the native merge step by step, including what happens to conflicting field values.
• How to undo a merge in Attio — what is recoverable after a merge, and for how long.
• Bulk merging duplicates in Attio · all comparisons · how Deduptio merges