Data model

Account hierarchy

Active Reach uses a four-layer account model. The 2026-05-26 revision made Location (outlet) a first-class attribute of a workspace rather than a separate tenant.


Organization (brand portfolio holder, usually one brand)
  └── Workspace (brand-tenant; one per brand)
        ├── Location[] (outlet — per-store ops, per-outlet overrides)
        └── Property[] (SDK install — web app / mobile app / kiosk)

Layer	Maps to	What lives here
Organization	Brand portfolio holder	Brand identity, billing entity, subscription, agency relationships
Workspace	Brand-tenant (one per brand)	Cross-outlet data: contacts, segments, channels, journeys, in-app, campaigns, ads, loyalty program
Location	An outlet (store)	Per-outlet overrides: local hours, manager phone, payment merchant, inventory, per-outlet channel sender mapping
Property	An SDK install (web app / mobile app / kiosk)	In-app + push surfaces, per-property metrics

workspace_id and location_id are distinct dimensions — never aliased. workspace_id is required on every tenant-scoped row; location_id is nullable (brand-tier rows = NULL; outlet-scoped rows stamp both).

Multi-brand operators run Agency mode: one Agency org → N Client orgs linked via agency_clients (1:N). The Agency org itself has no workspaces.

Customer URLs are subdomain-scoped to the brand with the outlet on the path: slug.actii.me/{outlet-code}/{page}. Admin URLs are path-scoped to the brand with the outlet in query state: actii.me/dashboard/{brand}/{page}?location=loc_X.

Five operational objects

Five objects form the backbone of campaign execution. Understanding their relationships helps you design integrations and debug issues.

Contact

A person your business communicates with.

Field	Description
`id`	System-generated unique ID
`user_id`	Your application’s user ID (set via `identify()`)
`anonymous_id`	SDK-generated ID (before identification)
`email`	Primary email address
`phone`	Primary phone number (E.164 format)
`properties`	Key-value pairs — standard (name, city, plan) and custom (any field you set)
`segments`	Array of segment IDs the contact currently belongs to
`created_at`	When the contact was first seen
`last_active_at`	When the contact last fired an event

Identity resolution: Contacts are deduplicated by user_id (highest priority), then email, then phone. When a match is found, records merge — properties combine (latest wins), event histories unify.

Event

Something a contact did.

Field	Description
`type`	`track`, `identify`, `page`, `group`, or `alias`
`event`	Event name (for `track` type, e.g., `order_completed`)
`user_id` / `anonymous_id`	Who did it
`properties`	Arbitrary key-value data describing the event
`context`	Device, browser, location, UTM metadata
`timestamp`	When it happened

Events are immutable — once ingested, they can’t be modified. They power segmentation, journey triggers, analytics, and attribution.

Campaign

A message sent to a segment through one or more channels.

Field	Description
`id`	Unique campaign ID
`name`	Human-readable name
`status`	draft, scheduled, sending, paused, completed
`segments`	Target audience (array of segment IDs)
`channels`	Which channels to use (email, whatsapp, sms, etc.)
`content`	Per-channel message content
`schedule`	When to send (immediate, scheduled, recurring, STO)
`conversion_goal`	Optional event name + tracking window for attribution

Journey

A multi-step automation defined as a directed acyclic graph of nodes and edges.

Field	Description
`id`	Unique journey ID
`name`	Human-readable name
`status`	draft, published, paused, completed
`nodes`	Array of node objects (trigger, wait, send, branch, split, exit)
`edges`	Array of connections between nodes
`version`	Incremented on each edit

See Journey graph schema for the full node/edge format.

Segment

A dynamic group of contacts defined by rules.

Field	Description
`id`	Unique segment ID
`name`	Human-readable name
`conditions`	Array of filter conditions (property checks, event checks, engagement checks)
`logic`	How conditions combine (`AND` or `OR`)
`member_count`	Current number of matching contacts (updated in real-time)

How they connect


Contact → fires Events → matches Segments → enters Journeys → receives Campaigns
                                    ↓
                              Analytics (aggregation, funnels, attribution)

Every downstream system (campaigns, journeys, analytics) reads from the same event stream. There’s no separate “marketing database” — events are the single source of truth.

What’s next

Journey graph schema — the canonical node/edge format
Event schema — the five event types in detail
Workspace / org boundary — in-tenant write-scope semantics and inheritance patterns