CTV was born identifier-poor, and the content metadata that reaches buyers is thin and inconsistent between apps. Content-audience mapping fills the gap: it profiles streamer domains, FAST channel sites and show pages in fixed, IAB Audience Taxonomy 1.1-aligned demographics, interests, purchase intent and personas. Honestly scoped for planning and curation — not impression-level bidstream classification.
On the open web you can at least argue about identifiers. In CTV the practical planning unit has always been the content: the app, the channel, the show.
No third-party cookies on a television, and device-level signals vary by platform. Cross-channel audience segments built for desktop translate poorly, leaving buyers reliant on platform-reported audiences they cannot inspect.
What one app calls a genre, another calls a category or omits entirely. Labels like “Lifestyle” span everything from budget DIY to luxury travel — content with visibly different audiences priced identically.
Every show, channel and streamer maintains a public web presence — official sites, episode pages, network domains. That footprint is classifiable with the same content-inferred method used on the open web, giving planners a neutral audience layer they can audit.
The output is a table your planning team already knows how to use: one row per channel, show or domain; audience attributes in fixed coded vocabularies as columns.
List streaming apps and their web domains, FAST channel lineups, network sites, and the show or franchise pages that anchor each property. Include publisher domains that carry the same editorial themes.
Join domains to the 102M-domain audience database for pre-computed attributes. For individual show or channel pages, request per-URL profiles through the real-time segmentation API — same vocabularies, page-level granularity.
Roll page- and domain-level attributes up to the unit you plan against — channel, show, genre block or daypart-equivalent content group. Keep INT.* and PI.* coded values so aggregations stay exact across quarters.
Weight spend across channels, brief CTV sellers with a shared audience language, shortlist content for sponsorships, and define curated packages where the content list — not a user identifier — is the targeting mechanism.
A planner is weighing a home-improvement FAST channel against a broader lifestyle network for a power-tools campaign. Both carry the same platform genre label. Profiling the channel's site and its flagship show pages produces the record shown here.
Precise brief match: home-ownership skew, INT.home_garden.home_improvement interest, and active purchase intent in PI.home_garden_services.home_improvement_and_repair.
Clear differentiation: the lifestyle network resolves to travel and dining attributes instead — a difference the genre label alone would never surface.
Fixed vocabularies: every attribute uses IAB Audience Taxonomy 1.1-aligned codes from the v1.0 controlled vocabulary.
Actionable weight shift: the coded diff gives a concrete reason to reallocate spend between the two channels.
INT.home_garden.home_improvementGardening INT.home_garden.gardeningPI.home_garden_services.home_improvement_and_repairHardware & Supplies PI.hardware_suppliesEach layer answers a different question. Content-audience mapping is the one a buyer can inspect independently, built from public content rather than platform-held user data.
| Approach | Granularity | Buyer can audit it | Depends on user-level data | Best used for |
|---|---|---|---|---|
| Platform first-party segments | Household / account | Platform-held | Yes | In-platform activation at scale |
| ACR / viewing-based data | Device viewing history | Vendor-held | Yes, opt-in dependent | Measurement and reach analysis |
| App / genre metadata | App or genre label | Visible in deals | No | Coarse inclusion / exclusion |
| Content-audience mapping | Domain, channel, show page | Yes — coded, versioned vocabularies | No — content-inferred, no PII | Planning, seller briefs, curation, sponsorship evaluation |
The same content-inferred method applied to the 40%+ of web traffic without third-party cookies.
Plan channels and content against 1,667 deterministic personas.
Assess whether a property's audience actually fits the brand before committing.
Content-to-audience logic for another identifier-poor channel.
How streaming platforms and FAST networks use the dataset on the sell side.
Every enumerated vocabulary value, aligned with IAB Audience Taxonomy 1.1.
No, and we are deliberate about that scope. The mapping works at the domain and content level — streamer and publisher web properties, FAST channel pages, show and episode pages — and produces planning-grade audience profiles. It is used before and after campaigns run: to build plans, curate deals and evaluate content. It is not an impression-level, pre-bid bidstream classifier.
Anything with a public content footprint: streaming services' and broadcasters' web domains, FAST channel and network sites, individual show, franchise and episode pages, and the publisher domains that syndicate or cover the same content. Domains come pre-computed in the 102M-domain database; specific show or channel pages are profiled per-URL through the real-time API.
Genre labels say what the content is; audience mapping says who it attracts. A home-improvement show and a luxury interior-design series can share a genre yet skew to different age brackets, income bands and purchase-intent segments. The mapping expresses those differences in fixed, IAB Audience Taxonomy 1.1-aligned vocabularies — 8 age brackets, 6 income bands, 285 sub-interests, 283 purchase-intent segments, 1,667 deterministic personas — each with a banded confidence value.
Self-serve tiers cover most planning teams: the top 100k domains with full audience attributes at $490 one-time ($190/quarter refresh), or the top 1M domains at $1,990 one-time ($590/quarter) with instant card checkout and immediate download. Vertical and country slices are $190–$490. Per-URL profiling of show and channel pages uses standard API plans, e.g. Pro at $99/month for 10,000 credits. Larger corpora up to the full 102M domains and custom feeds are quoted individually, with custom licensing from $15,000 per year. Full details on the pricing page.
Profile any channel site or show page in the demo and see the coded attributes a CTV planning matrix would carry — then license the domain corpus that matches your content universe.