An account list is a list of domains. Join it against a 102M-domain database of pre-computed audience segmentation and every account gains a structured profile — orientation, firmographics, interests, purchase intent and personas. Your ICP becomes a reproducible query across thousands of accounts, with no PII anywhere in the pipeline.
Each domain tagged with an orientation flag plus LinkedIn-standard firmographic bands for company size, seniority and job function.
All values drawn from versioned vocabularies aligned with IAB Audience Taxonomy 1.1 — reproducible and auditable.
Attributes are inferred from domain content alone. No cookies, no personal data, no consent overhead.
Most ICP documents live in slide decks. Expressed in enumerated vocabulary values, the same definition becomes a deterministic, reproducible query.
| ICP criterion (as written) | Dataset field | Example condition |
|---|---|---|
| "Sells to businesses, not consumers" | audience_type | audience_type = 'b2b' |
| "Mid-market, 50–1,000 employees" | b2b_company_size_employees | IN ('51_200','201_1000') |
| "Our buyers sit in data and engineering" | b2b_job_function | CONTAINS 'data_analytics' OR 'engineering_software' |
| "Decision-makers, not just practitioners" | b2b_seniority | CONTAINS 'director' OR 'vp' OR 'c_suite' |
| "Active in AI / modern data tooling" | interests | CONTAINS 'INT.tech_computing.artificial_intelligence' |
| "In-market for software" | purchase_intent | CONTAINS 'PI.software.computer_software' |
| "Only count solid signals" | confidence | IN ('medium','high') |
individual_contributor to c_suite, and 27 job functionsThe whole loop runs in a spreadsheet-plus-warehouse setup; nothing needs to touch your CRM until the final routing step.
Collect account websites from CRM records and prospect lists, plus account domains from work emails (drop free webmail). Reduce to a registrable domain — app.meridian-analytics.example becomes meridian-analytics.example — and deduplicate.
Match the list against the audience file with a left join on the domain key. Ad-hoc additions and long-tail misses can be resolved through the real-time API, so batch and API results mix cleanly.
Apply your ICP conditions as a weighted score: firmographic fit, function and seniority fit, topical fit via INT.*/PI.* codes. Keep weights in version control next to the vocab_version for full reproducibility.
Tier 1 goes to named-account plays and contact-level enrichment; Tier 2 to scaled nurture; the rest is deprioritized with a recorded reason. Feed tiers to your CRM and CDP, and rescore quarterly against the refreshed file.
A data-observability vendor scores a prospect from an inbound list. The joined row — real coded values from vocabulary v1.0 — settles tiering in seconds.
meridian-analytics.example
Tier 1 · ICP score 92
audience_type: b2b201_1000directormanagerdata_analyticsengineering_softwareINT.tech_computing.artificial_intelligencePI.software.computer_softwarePI.web_services.web_hosting_and_cloud_computingb2b201_1000data_analytics, engineering_softwaredirectorINT.tech_computing.artificial_intelligence, PI.software.computer_softwareEvery criterion clears at high confidence — the account routes to a named-account play. Only now does the team spend on contact-level enrichment. A b2c retail or 1_10-employee agency would be deprioritized before anyone booked a call.
Different instruments for different stages of the funnel — most ABM stacks end up using both.
| Question | Domain-level audience profiling | Contact-level sales intelligence |
|---|---|---|
| Unit of analysis | Registrable domain | Named person / org chart |
| Coverage | 102M domains, joined in batch | Companies and contacts in the vendor's graph |
| Personal data | None — content-derived, no PII | Contact records; consent and compliance overhead |
| Best stage | List scoring, ICP tiering, TAM mapping | Outreach to accounts that already cleared the ICP bar |
| Cost shape | Flat file tiers from $490; API credits for ad-hoc | Per-seat / per-record |
| Refresh | Quarterly file refresh; rescore is a re-run | Continuous, per-contact |
Scoring 20,000 raw accounts through per-record contact tools is expensive and unnecessary. Profile and tier on domains first, then spend contact-level budget only where the ICP score justifies it.
Agencies apply identical domain scoring to prospect research — see agency new-business research. Platform teams can license via the sales-intelligence & ABM industry page.
Attach account-domain profiles to CRM and CDP records as computed traits, with full schema and join-key mechanics.
Profile the domains referring traffic to your site — including the B2B sources sending you future pipeline.
Measure whether your ABM display spend actually delivered on B2B domains with the right functions and seniority.
Apply ICP-style domain scoring to prospective clients and their competitive sets before the first pitch.
One row per registrable domain with attributes from fixed, versioned vocabularies (v1.0): an audience_type flag (b2c / b2b / mixed), B2B firmographic bands aligned to LinkedIn-standard breaks — company-size by employees, seniority mix, and 27 job-function values — plus the audience character of the domain: demographics, INT.* interest codes (29 groups, 285 sub-interests), PI.* purchase-intent segments (34 groups, 283 segments), deterministic personas and a banded confidence value. Attributes are inferred from domain content; no PII is used anywhere in the pipeline.
Translate each ICP criterion into a condition on an enumerated field: company size becomes a set of b2b_company_size_employees bands (e.g. 51_200, 201_1000), buyer roles become b2b_job_function values (e.g. data_analytics, it_ops), market orientation becomes audience_type = 'b2b', and category relevance becomes INT.* or PI.* codes. Because every value comes from a fixed vocabulary, an ICP definition is a reproducible query, not a judgment call — the same list scores identically every time it runs. The complete value lists are on the taxonomy page.
Contact-level sales-intelligence tools resolve companies to named people, org charts and contact data. Domain profiling answers a different, earlier question: what kind of company is behind this domain and does its profile match our ICP — computed from the domain alone, across 102M domains, with no personal data involved. Many teams use both: domain profiling to score and tier the raw list at scale, contact-level tools only on the accounts that clear the ICP bar, where the per-record cost is justified.
Yes — that is what the database tiers are for. The Top 1M file ($1,990 one-time, instant download) covers most companies with meaningful web presence; vertical or country slices ($190–$490) suit focused territories; and for total-addressable-market builds against the full 102M-domain corpus, custom slices and feeds are quoted individually, with custom licensing from $15,000/year for OEM-scale use. Ad-hoc lookups for a few hundred accounts fit comfortably in an API plan instead. Compare options on the pricing page.
Look up any target-account domain in the demo dashboard, then license a database tier and tier your whole list in one join.