Every distribution thesis eventually reduces to one question no firmographic field encodes: does this company add value the customer would pay to keep — kitting, VMI, technical support, fabrication — or does it arbitrage a catalog?
Distributors answer that question on their own websites, in detail, every day. Screening the vertical means reading those answers at census scale.
Under one industry code sit three different businesses. Distributors publish the evidence that separates them — on services pages, line cards, and warehouse lists — but no screening tool reads any of it.
Kitting, VMI bins in customer plants, quality labs, fabrication — a sticky services business wearing a distribution costume. Evidence: services page, program language, certifications.
Wire-cutting operations, same-day delivery routes, territorial warehouses — economics driven by density. Evidence: fleet mentions, location pages, delivery radius claims.
Webshop, drop-ship, no warehouse — margin-thin and non-sticky. Evidence: “access to a million products” versus “30,000 SKUs in stock” tells the entire story.
Starting from the full classified web — not a directory — two passes separate signal from noise across the entire distribution-adjacent population.
Dead domains, directories, marketplaces, and retail discarded cheaply. In published runs, a 25,000-domain slice reduced to ~17,300 operating companies before the expensive pass touched anything.
Each surviving site read against your written thesis. Fifteen signals scored, every claim quoted with source URL. Group ownership zeroes the total and logs the evidence.
Kitting, assembly, fabrication, VMI, testing, e-procurement — quoted claims, not checkboxes
Principals carried, authorized-distributor language, buying-group memberships, franchise ties
Warehouse locations, square footage, delivery territories, fleet mentions
Evidenced from case studies and named accounts — aspirational lists filtered out
E-commerce depth, inventory visibility, punch-out and EDI capability
Founder/family language on about pages, or the footer that reveals a platform parent
From the standing 15-signal framework; these six carry distribution theses, each extracted with quoted evidence and source URLs.
The thesis-defining signal. We read services pages for kitting, fabrication, assembly, testing, and program management — quoted, not checkboxed — and classify the model: value-added distributor, route-density logistics, catalog reseller, or hybrid.
The line card is a distributor's balance sheet of relationships: manufacturers carried, authorized-distribution status, exclusive territories, buying-group memberships — captured verbatim, because authorized relationships transfer differently than gray-market sourcing.
Distribution's recurrence hides in program language: VMI, consignment stock, blanket orders, auto-replenishment portals. A distributor with VMI bins in customer plants has annuity characteristics a webshop never will — the evidence is public and we quote every instance.
Warehouse and territory footprint from the company's own locations pages: facility count, locations, delivery radius, with owned warehouses distinguished from agent mentions. Distribution economics are density economics, so footprint feeds both mandate fit and platform assembly math.
Acquired regional distributors keep their names for customer continuity, sometimes for a decade — the tell is a footer line, a “part of the family” page, or a harmonized privacy policy. We zero the score when found and quote the sentence; roughly one in ten keyword-perfect candidates fails here.
Founder and family association, stated: “family-owned since 1972,” second-generation leadership on the about page. Captured only as published and folded into Transition Context per our standards — roughly half of confirmed industrial fits carry this evidence, and distribution sits at the high end.
Composite reflecting real run behavior. Thesis: independent value-added distributors of fasteners and Class-C components, US, with VMI programs and quality capability, serving OEM customers.
Midwest distributor: VMI with barcode replenishment, kitting to work-cell sequence, in-house quality lab. Authorized relationships with major fastener principals. About page: “family-owned since 1972,” named second-generation president. Every element quoted, every quote sourced.
Homepage says “production components and inventory programs for OEMs” — no fastener keyword above the fold. Capabilities pages tell the real story. Profile databases file it as generic industrial supplies; a fifth or more of confirmed fits surface only this way.
No services, no territory, no programs — classified as catalog reseller, documented.
”Independent-looking” regional house — footer reads “a division of” a national platform. Sentence quoted.
Two thin sites — flagged rather than guessed. Extractable-signal count stated.
The screen's contract is coverage plus auditability — the full eligible universe, ranked on quoted evidence. These limits are stated per record, not hidden.
Revenue, gross margin, inventory turns, working-capital posture — nothing financial is website-visible, and we do not decorate guesses.
Whether any owner would take a meeting is unknowable from web evidence. Our standards prohibit pretending otherwise.
No ages, no personal speculation about founders — nothing beyond what the company states about itself.
Principals get added and dropped faster than sites update. Authorization claims deserve verification in diligence.
SKU counts and warehouse footage are the company's own figures. Customer names evidence a relationship's existence, not its size.
The quietest good distributors — running on relationships and a two-page site — land in the insufficient-evidence pile, honestly stated.
Scored CSV plus evidence appendix — model classification, line card, programs, footprint, ownership reads — quotes and URLs per row, CRM-ready.
Disagree with the top decile out loud. We convert disagreements into ICP terms and re-run the scored universe — included, and where theses get sharp.
Letters that cite a distributor's actual programs and line card read like a customer wrote them. Founders answer those and ignore the rest.
Annual monitoring re-reads the universe: new services pages, new principals on the line card, new footers with new parents. Deltas arrive flagged.
Leading company databases index the companies they found. We start from the entire active web and read it against your exact distribution thesis.
| Question | Code + keyword filtering | Full-web census screening |
|---|---|---|
| Value-add vs box-mover | Invisible — same code, same row | Classified from services-page evidence, quoted |
| The niche distributor in trade language | Missed — no category keyword | Read and classified; a fifth or more of confirmed fits surface this way |
| The acquired branch with a legacy site | Listed as independent | Ownership zero, footer quoted — about 1 in 10 keyword-perfect names |
| Line card and authorizations | Absent | Principal names and authorization language captured verbatim |
| VMI and program revenue | Absent | Program language quoted with source URLs |
| Thesis iteration | Re-filter, re-export, re-dedupe | Scored universe re-runs against your revised ICP, included |
| Auditability | Trust the filter | Verbatim evidence for every inclusion and every exclusion |
A distribution universe, once built, answers more than one question. The difference between theses is how the same evidence is weighted — every re-run is included, not sold twice.
Acquire a value-added anchor, tuck in regional houses. Heaviest weights: model classification and footprint. Anchor search screens for services depth and quality capability; tuck-in map weights territory adjacency and line-card overlap.
Manufacturer or larger distributor buying channel access. Heaviest weights: line card and authorized relationships. Program language — VMI, consignment, procurement integration — becomes the proxy for account stickiness.
Independent-sponsor play weighting transition context: founder association, operating history, visible bench — alongside a moderate value-add floor. Distribution suits this unusually well: ~50% carry explicit founder/family evidence.
Send the thesis — niche, geography, value-add requirements. A specimen arrives the same day; a scoped proposal follows.
Request the specimen report