Your ICP is a job function and a seniority band, not an age bracket. The Cookieless Audience database attaches B2B firmographics on LinkedIn-standard bands — job function, seniority, company size — to 102 million domains, so a SaaS marketing team can query “which trade media, practitioner blogs and community sites reach engineering directors at mid-market companies?” and get a ranked domain list back. No cookies, no identity graph: the profile is a property of the site itself, which matters because much of this niche practitioner content skews to browsers where third-party cookies are already blocked.
B2B SaaS demand generation leans on a small set of channels — LinkedIn, search, review platforms — because they are the places where a job title is known. But the buying committee spends most of its research time elsewhere: engineering blogs, ops newsletters, security communities, vertical trade press, comparison write-ups on independent practitioner sites. That inventory is cheap relative to its influence, and it is heavily cookieless: cookies remain on Chrome, but Safari, Firefox and iOS block third-party cookies by default, so roughly 40%+ of traffic carries no identifier your retargeting or intent vendor can match.
Domain-level firmographics invert the problem. Instead of asking “which known users can I follow?”, you ask “which properties concentrate my buyer function and seniority?” A Kubernetes tutorial site has an engineering-heavy, senior-skewing readership whether or not any visitor is identifiable — the audience is a stable property of the content.
Every attribute comes from a fixed, versioned v1.0 vocabulary aligned with IAB Audience Taxonomy 1.1, so a saved ICP filter means the same thing at the next quarterly refresh, and two marketers running it get the same list.
b2bengineering_softwaredirector / vp201_1000mediumFive filterable values instead of a persona paragraph. Run them over the domain file — or per-URL via the real-time API — and the output is a ranked list of sites where that reader concentrates.
The B2B layer sits alongside the full demographic, interest and intent model, so you can combine “who they are at work” with “what they read and what they’re shopping for” in one query. Full field reference on the audience segmentation taxonomy page.
Every domain carries an audience_type flag — b2c, b2b or mixed — so a SaaS plan filters out consumer-only inventory in the first pass.
27 function codes on a LinkedIn-standard scale: engineering_software, it_ops, security, data_analytics, product, marketing, finance, hr and more.
Seven bands from owner_founder and c_suite through vp, director and manager down to senior_ic and individual_contributor — match the level your deal actually closes at.
Six employee bands — 1_10, 11_50, 51_200, 201_1000, 1001_5000, 5001_plus — separating SMB content from enterprise trade press.
29 interest groups / 285 sub-interests (INT.tech_computing.artificial_intelligence) and 34 intent groups / 283 segments, e.g. PI.software.computer_software or PI.web_services.web_hosting_and_cloud_computing.
Every attribute ships with a banded confidence value — low / medium / high — so you can trade reach against certainty explicitly when the list feeds paid media versus outbound research.
No modelling step and no black box — the workflow is a sequence of filters over a flat domain file (or the same query via the API), and every intermediate state is a readable list of domains and codes.
Take the ICP one-pager — function, level, company size — and restate it in vocabulary terms rather than prose.
“Security leaders at mid-market companies” becomes security + director/vp + 201_1000.
Apply the filters across 102M domains with audience_type = b2b/mixed and a medium+ confidence floor.
Sort by how strongly the function and seniority over-index versus the corpus baseline; sanity-check the head of the list.
Feed the ranked list into sponsorships, newsletter buys, content syndication, PMP/Deal-ID curation or ABM allow-lists.
A DevOps observability vendor sells to engineering and platform teams at companies of 200–5,000 employees, with directors and VPs signing. Here is that ICP as a database filter — codes on the left, what they read as on the right.
| Field | Filter value (code) | Reads as |
|---|---|---|
| audience_type | b2b or mixed | Business-facing readership |
| b2b_job_function | engineering_software or it_ops | Software engineering / IT operations |
| b2b_seniority | director or vp | Director-to-VP decision makers |
| b2b_company_size_employees | 201_1000 or 1001_5000 | Mid-market to lower enterprise |
| interest | INT.tech_computing.computing | Computing content |
| purchase intent | PI.web_services.web_hosting_and_cloud_computing | Hosting & cloud computing buyers |
| confidence | medium or high | Well-evidenced attributes only |
Not a replacement for your ABM platform or LinkedIn budget — a complementary instrument that covers the open-web layer those channels don’t describe.
| Walled-garden B2B ads | Account-level intent data | Domain-level firmographics (this dataset) | |
|---|---|---|---|
| What it describes | Known member profiles inside one platform | Research activity of target accounts | The audience of every content site your buyers read |
| Open-web coverage | None — inventory stays in-platform | Limited to the vendor’s publisher co-op | 102M domains, head to long tail |
| Cookieless traffic | Logged-in, so unaffected but capped in reach | Often identifier-dependent | Fully covered — profile is a property of the domain |
| Granularity | User-level, single platform | Account-level topics | Function × seniority × company size per domain, plus interests and intent |
| Reproducibility | Platform-defined segments | Vendor-defined topic taxonomies | Fixed v1.0 vocabularies; same filter, same meaning |
The firmographic attributes describe the site’s readership, not any individual visitor. A cloud-infrastructure blog concentrates software-engineering readers at senior levels; a small-business bookkeeping guide concentrates owner/founders at 1–10-employee companies. Those are stable, content-derived properties of the domain, expressed on LinkedIn-standard bands with a banded confidence value per attribute. No individual is profiled and no PII enters the pipeline.
No — and it is not trying to. ABM platforms activate against known accounts and contacts; this dataset tells you which media properties concentrate your buyer function, seniority and company-size band, so you can plan sponsorships, curate PMP deals, build allow-lists and give your ABM programme open-web air cover. Pair it with the ABM account profiling workflow when you want firmographic context on referring or researched domains.
Technical and professional audiences over-index on browsers and devices where third-party cookies are blocked — Safari, Firefox, iOS — and cookies remaining on Chrome doesn’t help you on the rest, which is roughly 40%+ of traffic. Practitioner blogs and niche trade media are precisely the inventory retargeting under-counts. Domain-level firmographics describe that inventory with the same fields and confidence bands as everything else, because they never depended on an identifier.
The top 100k domains with full audience attributes cost $490 one-time (or $190/quarter refresh); the top 1M domains cost $1,990 one-time ($590/quarter) with instant card checkout and immediate download. Vertical and country slices run $190–$490. Larger cuts — 5M up to the full 102M corpus, custom enrichment or feeds — are quoted individually. API plans (e.g. Pro at $99/month for 10,000 credits) cover per-URL lookups. See pricing.
Firmographic context on the domains your target accounts read and run.
Plan against 1,667 deterministic personas across 102M domains.
Compare your media footprint against the sites competitors court.
How ABM and sales-intelligence platforms use the same firmographic layer.
Filter live firmographic data in the demo dashboard, or download the top-1M domain file and have a ranked buyer-media list before your next planning cycle.
Open the audience demo See database pricing