Every row in the delivery and in the free 1,000-lead sample is the same deep object: the owner, their company, its measured website tier, an explainable score, and verified contacts — not a name-and-email pair, a case file. This page shows the object exactly as it ships: layer by layer, then a real (masked) example, then every field group.
the object below is a real pull · identity masked for publication · the delivery and the sample ship it unmasked · MEASURED · Outricher data
The object is assembled in three passes. Each layer is independently verifiable — nothing in a later layer overwrites the raw fields of an earlier one.
One row from the shipped delivery — a janitorial-services owner, tier A+, score 105 (the pool maximum). Identity and contacts masked for publication; the delivery and the free sample ship this object unmasked. Abridged for the page — the shipped row carries 133 filled fields.
{
"formatted_name": "R███ P█████", // masked for publication
"headline": "Janitorial",
"position_title": "Small Business Owner",
"title_class": "primary_owner",
"location_name": "⟨city⟩, Ohio, United States",
"location_country_code": "US",
"public_profile_url": "⟨professional profile URL⟩",
// ── layer 2: the business ─────────────────────────
"company_name": "⟨janitorial services co⟩",
"company_industry": "Facilities Services",
"company_employee_count": 6,
"company_founded": 2010,
"company_locality": "Houston, Tx",
"company_region_code": "TX",
"company_trade_category": "cleaning",
"company_trade_match": "name", // trade proven by the company name itself
"company_website_tier": "none", // ← the sales signal: no website at all
"company_website": null,
"company_residential_focus": true,
"company_follower_count": 29,
// ── layer 3: score — every point named ────────────
"combined_score": 105,
"tier": "A+",
"company_score_details": [
"fit_sweet_spot_size +12",
"fit_trade_in_name +14",
"fit_residential +5",
"fit_local_presence +3",
"fit_established +3",
"web_no_site_hot +20", // the inverted axis: no site = hottest
"reach_active_company_page +3",
"fit_high_density_metro +3"
],
"lead_score_details": [
"fit_decision_maker +18",
"reach_email +10",
"reach_phone +5",
"reach_socials +5",
"fit_trade_in_headline +4"
],
"batch_codes": ["B2", "B7"], // build-from-scratch + cleaning-trade slices
// ── contacts: verified-or-empty ───────────────────
"email": "p███████@yahoo.com",
"phone": "+1937███████",
"db_personal_emails": ["█████ 4 verified addresses █████"],
"db_personal_numbers": ["█████ 1 number █████"],
"db_facebook_url": "⟨social profile⟩",
// ── freshness & provenance ────────────────────────
"db_profile_updated_at": "2026-06-26T07:52:39Z",
"api_about": "⟨the owner's own 60-word service description — ships in full⟩",
"api_experience": { /* dated career history where the profile carries it */ }
}web_no_site_hot +20 is not a black-box weight — it says exactly why this lead is in batch B2: a residential cleaning business in Houston, six people, founded 2010, with an active company page and no website. Your first sentence to them writes itself from the record.The same fields, four different plays. Every filter below is a column in the CSV — no re-processing needed.
company_website_tier ∈ {none, social_as_site, free_builder} → the need is measured, the offer is obvious. 275 such rows in the free sample; batch B2 in full.company_website_tier = own_domain + low company_follower_count → a domain exists but the presence is dormant. Batch B3 — the best contacts in the pool (email 100%, phone 50%).company_trade_category — 12 trades, classified from names and descriptions, not generic industry labels. Batches B4-B7 pre-cut the four biggest.phone non-empty (52% of the sample) + sort by combined_score → owner-operators pick up their own phones.email or phone is a promise, not a failure — nothing is guessed or pattern-generated. Coverage: email 85% of the delivered pool (100% in the sample), phone 33% (52% in the sample).company_region_code resolves on ~61% of rows (small businesses often list no structured address); api_about exists where the owner wrote one (~32%). Sparse fields ship as null — they are never backfilled.179 distinct fields across the pool; a typical row carries 134. Grouped as they appear in sample-1000.jsonl.
The 1,000-lead sample ships this exact object, unmasked, with CSV and lossless JSONL. The full 1,583-lead delivery is free on request — and the same pipeline cuts a fresh audience for your offer. More about us: outricher.com/intro.