Safety tools tell you where a brand must never run. They say nothing about where it belongs. Fit analysis matches your target profile against each domain’s coded audience attributes across 102M domains, producing a transparent score from readable fields — not an opaque vendor rating.
Most brand-protection budgets stop at the first rung of a three-rung ladder. Each rung asks a progressively more valuable question, and only the third requires audience data.
Binary exclusion of illegal, hateful or dangerous content. Universal across brands, handled by blocklists. Necessary — and entirely negative: removes the worst without rating the rest.
Brand-specific content judgments — an airline’s view of aviation-incident news differs from a newspaper’s. Content-category data from our API drives this rung. Still silent on who is reading.
A domain can be safe, suitable — and read by entirely the wrong people. Fit compares your target profile with each domain’s coded audience, turning “acceptable inventory” into “right inventory.”
Simple arithmetic over coded fields — documented on the taxonomy page — so any score can be challenged, decomposed and reproduced by another analyst.
Field-by-field agreement on age brackets, gender skew, income band, life stage and urbanicity between the domain’s profile and your target.
Shared INT.* codes between the brand’s interest set and the domain’s — weighted toward specific sub-interests over broad tier-1 groups.
Shared PI.* segments — the strongest fit evidence, because purchase intent is the commercial behavior the brand is buying access to.
Each component is weighted by low / medium / high confidence bands. A strong match on weak evidence never outranks a moderate match on strong evidence.
Five steps from brand profile to a scored, tiered domain list — a spreadsheet exercise on the database file, or a scripted job via the API.
Write the target audience in vocabulary codes: demographics, interests, intent segments — the same fixed values the database uses.
Start from the media plan, a curated package, or a database tier — every candidate already carries its full attribute set.
Compute demographic match, INT.* overlap and PI.* alignment per domain, each weighted by confidence bands.
Bucket domains into strong fit, partial fit and mismatch. Components tell you exactly why each domain landed where it did.
Shift budget toward strong-fit domains, price partial fits accordingly, and cut mismatches that safety tools would have approved.
The brand’s encoded target profile, then three candidate domains scored against it:
25_34–45_54, balanced gender, upper_middle incomeINT.travel.adventure_travel, INT.travel.camping, INT.healthy_living.fitness_and_exercisePI.sporting_goods.outdoor_recreation_equipment, PI.travel.camping| Candidate domain (archetype) | Demographics | Interest overlap | Intent alignment | Fit verdict |
|---|---|---|---|---|
| Adventure-travel & trail blog 35–44, balanced, upper-middle, high confidence |
Match on age, gender, income | adventure_travel, camping, fitness_and_exercise all present |
outdoor_recreation_equipment + PI.travel.camping, high confidence |
Strong fit |
| Metro news site Broad 25–64, mixed income, medium confidence |
Partial — audience broader than target | Weak — news & politics dominate; no camping or adventure codes | None of the target PI.* segments |
Partial fit |
| Esports team-coverage site 18–24, male-strong, high confidence |
Mismatch on age and gender skew | INT.video_gaming.esports — no overlap with target set |
Gaming and electronics intent, not outdoor | Audience mismatch |
| Safety blocklist | Suitability categories | Audience fit (this dataset) | |
|---|---|---|---|
| Question answered | Where can we never run? | What content matches our standards? | Does the readership match our buyer? |
| Basis | Content exclusion lists | Content categories per brand policy | Coded audience attributes: demographics, INT.*, PI.*, personas |
| Output | Binary block / allow | Category-level allow / avoid | Graded fit score with inspectable components |
| Budget effect | Removes the worst | Shapes the middle | Concentrates spend on the best |
| Relationship | Complementary layers — fit analysis assumes safety and suitability screening remain in place and adds the audience dimension on top | ||
Safety excludes unacceptable content universally. Suitability applies brand-specific content judgment. Fit is an audience question — do the domain’s readers resemble the brand’s buyers? A domain can pass the first two completely yet fail on fit. The three layers are complementary, not interchangeable.
No — deliberately. The database ships raw coded attributes (demographics, interests, intent, personas, confidence bands). You compute fit against your brand profile with transparent arithmetic. Scores stay decomposable, and different brands — or different campaigns — can score the same domain differently.
Yes. The audience profile is a property of the domain, not of visitor identifiers. Domains whose traffic skews to Safari, Firefox or iOS — roughly 40%+ of traffic today — are scored exactly like everything else. Fit analysis never depended on cookies.
For campaign-level fit scoring, the database file is enough — top-100k at $490 one-time, top-1M at $1,990 with instant download. Quarterly refreshes keep scores current. The real-time API adds page-level segmentation for auditing specific URLs rather than whole domains.
Build the plan that fit scoring then stress-tests.
Bake fit criteria into the Deal-ID inclusion list itself.
The same fit logic applied to sponsorship and partnership properties.
Score affiliate candidates for audience match before commission talks.
Encode your target profile, score your current plan, and see which approved domains are quietly buying the wrong audience.
Open the audience demo See database pricing