Data API

Data Model

How the four Parcel datasets relate to each other.

The Data API exposes four datasets. All record ids are UUIDs.

The four datasets

Dataset	What it represents
`accounts`	Organizations: developers, architects, engineers, GCs, agencies, law firms, consultants, and others active in the construction pipeline.
`contacts`	People at accounts. Each contact has an `account_id` linking it to its parent account.
`projects`	Individual construction or development projects tracked through the permitting and entitlement lifecycle.
`signals`	Timeline events on a project (permits filed, approvals, team updates, etc.). Each signal has a `project_id`.

Relationships

accounts ──< contacts          (one account, many contacts)
projects ──< signals           (one project, many signals)
accounts >──< projects         (many-to-many via project_accounts, with a role)
contacts >──< projects         (many-to-many via project_contacts, with a role)

Relationship	How it works
Contact → Account	`contacts.account_id` is a direct FK to `accounts.id`.
Account ↔ Project	Joined through `project_accounts`. Each row carries a `role` (e.g., developer, architect).
Contact ↔ Project	Joined through `project_contacts`. Each row carries a `role`.
Signal → Project	`signals.project_id` is a direct FK to `projects.id`.

Convenience count fields

Each dataset includes pre-computed *_count fields so you can gauge relationship depth without a separate query:

Dataset	Count fields
`accounts`	`projects_count`, `contacts_count`
`contacts`	`projects_count`
`projects`	`accounts_count`, `contacts_count`, `signals_count`

Signals do not carry count fields.

Traversing relationships

Use the include parameter on GET /data/{dataset}/{id} to embed related records directly in the response. For example, fetching a project with include=accounts&include=signals returns the project record with its related accounts array and signals array nested under those keys.

See Related Records for the full include reference per dataset.