# Contacts \[Name the addresses a team transacts with]

A contact is a team-wide name for an external address. Once saved, the name stands in for the raw address everywhere addresses appear: recipient search in [sends](/transactions/sends), transaction feeds, the [Accounting page](/accounting), and CSV exports. Contacts are shared by the whole team, and any member can create, edit, or delete them.

## Creating

Add contacts from the *Contacts* page, inline from any address shown in a transaction feed, or from the command menu. A contact is an address plus a display name; you can enter an ENS name, which is resolved once and stored as its address. Each address holds one contact per team, applied across all networks.

Two restrictions: contacts must be EVM addresses, and the team's own Splits accounts can't be added as contacts (they're already named).

## How addresses get named

When Splits displays an address, it uses the first name it finds: the team's account and bank-account names, then contact labels, then live ENS and Farcaster lookups, then [payee](/contacts/compliance) details, and finally the shortened address. Saving a contact pins a name the whole team sees, instead of depending on what an address resolves to.

## Other named recipients

Two recipient types are stored separately from contacts but appear alongside them in recipient search:

* **Payees**: counterparties with a verified identity and tax forms; see [Compliance](/contacts/compliance).
* **Vendor bank accounts**: fiat payees with a deposit address; see [Paying vendors](/banking/paying-vendors).

## Programmatic access

Via the [Splits CLI / MCP](/introduction/agents):

* `splits contacts list --q "acme"`: search contacts by name or address; returns up to 500 (**Read** scope)
* `splits contacts lookup --addresses 0x...,0x...`: batch address-to-name lookup, max 100 addresses (**Read** scope)

Creating and editing contacts is web-only today.