Categorisation Rules

Categorisation rules automate the classification of bank import transactions. When a transaction's description, reference, amount, or contact matches a rule, the matching rule's target ledger account is suggested (or auto-assigned) for that transaction.

Rule sources and precedence

Three sources of rule apply, in this order of precedence:

Your own rules, created for your organisation, take priority. Seed rules, provided when your organisation is set up, apply next. Shared platform defaults, derived across many organisations, apply last. So a rule you create always wins over a seed rule or a platform default for the same transaction.

When more than one rule matches a transaction, the outcome is resolved deterministically: more specific operators and fields win first, then your assigned priority, then the earliest rule. The test endpoint returns a diagnostic describing why a given rule won.

Fields and operators

A rule matches one transaction field with one operator. The fields are description, contact_name, amount, reference, and the metadata keys metadata.category, metadata.project, metadata.department, and metadata.tag. The operators are contains, equals, starts_with, ends_with, greater_than, less_than, and regex.

→ The Categorisation Rules object

Endpoints