Whether you sell freight services, source competitive intelligence, or hunt distribution partners, the question is the same: who actually imports this product into the US? The answer is not in a company directory. It is in customs records — specifically, in the bills of lading filed for every ocean container that enters the country. This guide walks through the full method: how HS codes work for prospecting, when to search by product keyword versus HS chapter, how to read an individual BOL record, how to size an importer by TEU and shipment cadence, and how to turn the raw results into a working target list.
HS codes: the addressing system for finding importers of a product
Demo on your top 3 lanes — we'll show you the saved-company workflow against your actual book of business.
The Harmonized System is a numeric taxonomy for everything that crosses a border, maintained by the World Customs Organization and used by nearly every country. It nests: the first two digits are the chapter (94 = furniture), four digits are the heading (9403 = other furniture and parts), and six digits are the subheading (9403.60 = wooden furniture other than office or kitchen). The first six digits are identical worldwide; the US extends them to a 10-digit HTS code for tariff purposes, but for prospecting you almost never need to go past six.
A few working examples: coffee is 0901, auto parts cluster in 8708, plastic articles in 3926, lithium-ion batteries in 8507.60, bicycles in 8712. Spend ten minutes with a free HTS lookup and note the two or three headings that cover your target product — that short list is the backbone of every search you run afterward. The practical rule: chapter level (2 digits) is for market sizing, heading level (4 digits) is for building prospect lists, and subheading level (6 digits) is for separating close cousins, like distinguishing wooden furniture importers from metal-furniture importers inside 9403.
Product keyword vs HS chapter search: when to use which
BOL cargo descriptions are free text typed by forwarders and carriers, which makes keyword search precise but leaky. A search for "sofa" will hit records described as SOFA, SOFAS, and SOFA SET — and miss the same freight described as SEATING, UPHOLSTERED FURNITURE, or a pure HS number with no words at all. Misspellings are common, and about a fifth of descriptions are too generic to match on. Keyword search finds high-confidence importers fast; it never finds all of them.
HS-based search has the opposite profile: it catches everything classified under a heading regardless of how the description was worded, but the heading is broader than your target — 9403 includes office furniture, hospital beds, and shop fittings alongside the living-room sofas you care about. The working method is to combine them: start at the heading level for coverage, then apply keyword filters inside the results for precision. Run the keyword-only search too and dedupe; the delta between the two result sets is usually where the overlooked, under-prospected importers live.
A worked example, end to end. Say you broker furniture freight and want every US sofa importer. Start with heading 9403 plus 9401 (seats), pull twelve months of records, and keyword-filter the results for SOFA, COUCH, SECTIONAL, and UPHOLSTER. Then run the standalone keyword search and merge the two sets. In a typical pull, the HS-first pass finds the large importers everyone already knows, while the keyword-only delta surfaces smaller consignees whose records were classified oddly — the accounts no competitor is calling. Fifteen minutes of query work, and you are holding a candidate list no directory sells.
How to read a BOL record
Every record in a bill of lading search carries the same core fields. The consignee is the US party receiving the goods — usually the importer, which means usually your prospect. The shipper is the overseas supplier, which tells you the sourcing country and factory relationships. Then: the carrier and vessel, the port of lading (origin) and port of unlading (US entry), the cargo description and HS code, container numbers and counts, gross weight, and the arrival date. One record is an anecdote; twelve months of records for the same consignee is an operating profile.
Two traps to avoid when reading records. First, the consignee is sometimes a freight forwarder, an NVOCC, or a customs broker rather than the true importer — if the consignee name is itself a logistics company, check the notify party field, which often names the real buyer, and filter known logistics entities out of your list. Second, some records say "TO ORDER" or carry a redacted name because the importer requested confidentiality from CBP; a good data platform reconstructs many of these through address matching and historical linkage, but expect some blanks in any product category.
Sizing an importer by TEU and shipment cadence
Volume tells you deal size. One TEU is one twenty-foot-container equivalent (a standard 40-foot box counts as two), and annual TEU is the cleanest single sizing metric. Rough bands: under 25 TEU a year is a small importer — real business, small freight wallet; 25 to 100 TEU is a solid SMB account, often with nobody dedicated to logistics; 100 to 500 TEU is a mid-market account with a logistics manager and meaningful spend across ocean, drayage, and domestic legs; 500-plus TEU is enterprise territory with formal RFPs. Match the band to what you sell before you write a single email.
A quick revenue heuristic for freight sellers: once ocean, drayage, and fees are counted, a mid-market importer's international spend runs very roughly $2,000 to $4,000 per TEU landed, so a 200-TEU account represents somewhere around half a million dollars of annual freight wallet before the domestic legs are added. Precision does not matter at the list-building stage; triage does. Size the wallet before you spend a sequence on it, and let the small accounts wait their turn.
Cadence tells you timing. Weekly or biweekly arrivals signal replenishment-driven supply chains and contract freight. Strong seasonality — toys and decor peaking in Q3 arrivals ahead of holiday retail, garden goods landing in Q1 — means you should be in the inbox four to five months before peak, because bookings happen roughly 90 days ahead of arrival. And a growth screen is the best prioritization filter there is: an importer whose trailing-90-day TEU is up 30 percent or more year over year is adding suppliers, lanes, and capacity needs right now. Sort your list by that one signal and the top decile earns the first calls.
Weight and container counts also catch what TEU alone misses. An importer landing 40 light-but-bulky containers of outdoor cushions is a different freight buyer than one landing 40 heavy containers of ceramic tile at max payload — same TEU, different equipment needs, different drayage economics, different pitch. Skim gross weights across an account's records before the first call; when weight per container runs high, payload limits and chassis planning become the conversation, and the rep who raises them first sounds like an operator rather than a vendor.
Building the importer list, step by step
Pulling the method together into a repeatable build:
- Define the product footprint. Two or three HS headings plus a handful of keyword variants, including the misspellings you saw in real records.
- Pull 12 months of BOLs. A full year smooths seasonality; a 90-day pull will systematically miss seasonal importers.
- Clean the consignees. Dedupe name variants of the same company, strip forwarders and NVOCCs, and resolve "TO ORDER" records where the platform can.
- Segment by TEU band and cadence. Tag each account small/mid/large and steady/seasonal/lumpy — those two tags determine both the pitch and the send date.
- Enrich and prioritize. Attach verified logistics contacts, then sort by growth signals — volume acceleration, new lanes, new suppliers — so the accounts in motion get touched first.
Done manually against raw customs files, this build takes days of cleanup before the first email goes out — entity resolution alone is a project. An importer database that has already deduped consignees, mapped HS codes, computed TEU histories, and attached contacts collapses the same build into about an hour, which is the difference between running this once and running it as a monthly motion.
From product to pipeline
Finding the importers of a product is a query, not a research project: two HS headings, a keyword filter, twelve months of BOL data, a TEU sort, and a growth screen. Everything downstream — the outreach, the meetings, the booked freight — depends on how well that query is built. If you want to skip the raw-data cleanup and start from a clean, contact-enriched list, start with importer leads — search your product, see every US importer moving it, and export the list the same afternoon.