Data model

How everything in Profound nests together — from your Organization down to the slice dimensions on every report.

Your Organization holds many Categories. Each Category contains its Assets (the brands you’re tracking), grouped into Topics, made up of Prompts. Each Prompt then fans out across four dimensions — Platform, Region, Persona, and Tag — which is how reports slice the data.

Entity reference

Entity	What it is
Organization	Your Profound account. Has many Categories.
Category	A workspace — typically one brand and its competitive set. Every report query is scoped to one Category.
Asset	A brand, company, or product tracked in the Category. Has is_owned, website, and alternate_domains.
Topic	A grouping of related prompts (e.g. “Pricing”, “Integrations”).
Prompt	A question collected against AI models (e.g. “Best CRM for startups?”).
Platform	The AI surface that answered (ChatGPT, Claude, Perplexity, …). Called model in API params.
Region	The country the answer is scoped to (US, UK, Germany, …).
Persona	An audience profile (e.g. “small business owner”).
Tag	A free-form label applied to a Prompt.

What this means for queries

1. Every report call is scoped to one category_id. No cross-category queries.
2. filters narrow which prompts contribute — by asset_name, topic_id, tag_id, and the like.
3. dimensions break results out by slice — model, region, persona, or tag.

Endpoints by entity

Entity	List endpoint
Categories	GET /v1/org/categories
Topics	GET /v1/org/categories/{id}/topics
Prompts	GET /v1/org/categories/{id}/prompts
Tags	GET /v1/org/categories/{id}/tags
Assets	GET /v1/org/categories/{id}/assets
Personas	GET /v1/org/categories/{id}/personas
Platforms (org-level)	GET /v1/org/models
Regions (org-level)	GET /v1/org/regions

Click any endpoint above to jump to its full parameter reference.