Persona-level audience planning used to require the kind of enterprise data contract only a holding company signs. Cookieless Audience puts the same class of instrument on a planner’s desk for $490: a 102M-domain database with pre-computed demographics, 285 sub-interests, 283 purchase-intent segments and 1,667 deterministic personas — every attribute from fixed v1.0 vocabularies aligned with IAB Audience Taxonomy 1.1. Plan by persona, research a pitch overnight, and select campaign inventory that covers the roughly 40%+ of traffic cookies never see.
Independent and mid-size agencies live in a gap. Enterprise planning suites are priced and contracted for the largest buyers — annual commitments, per-seat licences, procurement cycles longer than most campaigns. DSP-native segments are free but opaque: vendor-defined, cookie-dependent, and impossible to defend in a client meeting when someone asks what “Auto Intenders” actually means.
This dataset takes the opposite position on every axis. It is self-serve — the top-100k file is $490 one-time, the top-1M file is $1,990 with instant card checkout and immediate download. It is transparent — every attribute comes from a fixed, published v1.0 vocabulary you can hand to a client’s analytics team. And it is identifier-independent: the audience profile is a property of the domain, so it describes Safari, Firefox and iOS inventory exactly as well as Chrome. Cookies remain on Chrome, but roughly 40%+ of traffic is already cookieless, and privacy regulation keeps adding pressure on identifier-based planning.
The practical effect: a two-person planning team can run the same class of persona analysis a data-science department would — in a spreadsheet, this afternoon.
Vertical and country slices run $190–$490 when a single client category is all you need. Larger corpus cuts are quoted individually.
One dataset, three points in the agency cycle: winning the account, planning the buy, and selecting the inventory.
Before the pitch, profile the prospect’s category: which domains their buyers concentrate on, which INT.* and PI.* codes define the audience, and where competitors’ likely media footprints sit. Walk in with an audience map the incumbent doesn’t have. See agency new-business research.
Restate the brief in vocabulary terms — age bracket, life stage, income band, interests, intent — and filter 102M domains down to the properties where that persona over-indexes. Reproducible across planners, offices and quarters. See media planning by persona.
Turn the ranked domain list into activation: allow-lists for open exchange buys, curated domain sets for PMP and Deal-ID packages, and direct-buy shortlists — including the long-tail and Safari-heavy properties cookie-based tools under-count. See inventory curation.
Every intermediate state is a readable list of domains and codes — no black box between the brief and the buy. The same filters run per-URL through the real-time API when you need page-level granularity for analysis.
Restate the client’s target description as coded attributes from the taxonomy: demographics, interests, purchase intent.
Apply the codes to the domain file, with a confidence floor (medium+) to keep only well-evidenced matches.
Sort matches by how strongly the target persona concentrates on each domain versus the corpus baseline.
Cross-check brand fit, competitor overlap and cookieless share so the plan holds up in the room.
Export allow-lists and PMP domain sets; re-run the identical filter at the quarterly refresh to keep plans current.
An independent agency is pitching a direct-to-consumer dog-food brand. Before the first meeting, a planner builds the category’s audience map. Here is the brief’s target — “suburban dog owners, families, mid-to-upper income, actively buying for their pets” — as a database filter.
| Field | Filter value (code) | Reads as |
|---|---|---|
| age_bracket | 25_34, 35_44 | 25–44 year olds |
| life_stage | family_young_children | Families with young children |
| income_level | middle, upper_middle | Mid-to-upper income |
| urbanicity | suburban | Suburban households |
| interest | INT.pets.dogs | Dogs |
| purchase intent | PI.pet_services.pet_stores | Pet stores & supplies |
| purchase intent | PI.pet_services.veterinary_services | Veterinary services |
| confidence | medium, high | Well-evidenced attributes only |
| Enterprise planning suites | DSP-native segments | Panel-based tools | Domain-level audience file (this dataset) | |
|---|---|---|---|---|
| Commercial model | Annual contract, per-seat licences | Bundled with media spend | Subscription | From $490 one-time, self-serve |
| Segment transparency | Varies by vendor | Vendor-defined, opaque | Methodology published, coarse | Fixed v1.0 vocabularies, fully published |
| Cookieless coverage | Varies | Weak where cookies are blocked | Panel-modelled | Full — profile is a property of the domain |
| Long-tail domains | Head-of-market focus | Sparse ID matches | Mostly absent | 102M domains, head to tail |
| Client defensibility | Contractual restrictions on sharing | “Trust the platform” | Accepted currency, limited depth | Codes any client analyst can verify |
Yes — the database tiers are licensed to your organisation, so the same file can support planning and research across your client roster. If you need feeds, OEM redistribution, or client-facing products built on the data, that falls under custom licensing (from $15,000/year) — contact us to scope it.
DSP-native segments are typically cookie- or ID-derived, vendor-defined and opaque — you cannot see why a user is in “Auto Intenders”, and coverage collapses on Safari, Firefox and iOS. This dataset works at the domain level with fully published vocabularies: you can show a client exactly which coded attributes selected each domain, and the profile covers cookieless inventory identically because it never depended on an identifier.
The output of a persona filter is a ranked domain list — which activates as an allow-list, a curated PMP or Deal-ID package, or a direct-buy shortlist in whatever platform you trade through. The dataset does not do impression-level pre-bid classification in the bidstream, and we say so plainly; it is the planning and curation layer that decides which inventory is worth bidding on at all.
Most planning teams start with the top-100k file ($490 one-time, $190/quarter refresh) or a $190–$490 vertical slice for a single pitch. Teams that want the long tail take the top-1M file at $1,990 one-time ($590/quarter) with instant checkout. Per-URL analysis via the API starts at $99/month for 10,000 credits — see pricing.
The core planning workflow, step by step.
Build the audience map before the pitch meeting.
Compare category footprints with the same coded fields.
The sell-side view: documented audiences and SDA labels.
Filter live audience data in the demo, or buy a database tier with a card and have the file open in a spreadsheet within the hour.
Open the audience demo See database pricing