All topics
Get started8Account, team & permissions44Units & fields70Records, views & tasks39WhatsApp & Cortex Chat68Email & social27Lead ads & distribution38Automations52Documents & quotes32Dashboards40Cortex AI18Developers & API12
Help Center·448 guidesAll topicsContact support

How Corteksa spots a duplicate: match keys and normalization

The rule compares the match fields after normalizing their formats: +966, 00966 and 966 are one number, and email ignores upper and lower case.

4 min

Deduplication control is a paid Add-on.

Any key or all keys

Any key matches: one matching field is enough. All keys match: every filled-in field must match.

A field that’s empty in the incoming data isn’t used for matching.

Arriving at the same time

If three messages or three form submissions from the same client arrive at the same moment, they’re handled one at a time, so only one record is created.

In Unify mode, the target is the matching record with the most recent activity.

A client whose record was moved

If you moved a client’s record to another unit and they then message you, their chat is linked to their new record instead of creating a new one. See Move a record from one unit to another.

At a glance

Field type
How it’s normalized before comparing
Phone and WhatsApp number
Symbols are removed so only the digits are left, and a leading 00 is removed
Email
Spaces are removed, and upper and lower case count as the same
Provider ID
Spaces are removed, and case matters
Other types
Extra spaces are removed, and case doesn’t matter

Good to know

Matching only looks at records that aren’t deleted.
A unique field compares the exact text, while matching here happens after normalizing.
Did this guide solve it?
If not, we’ll ask what was missing.
Open in Corteksa
Did this guide solve it?

We read every answer. To get a reply, use Contact support.