B2B market intelligence: map the buyers your channels never showed you
Revenue becomes predictable when a company maps the buyers its channels never showed it and engineers its pipeline from data it can verify. This guide shows how a mid-market manufacturer or distributor does that with public records, and where purchased contact databases stop.
- What is B2B market intelligence, and why should a mid-market manufacturer care?
- Why do trade shows, distributors and inbound leave buyers unseen?
- Which public records show the buyers your channels missed?
- How do you size your real addressable market from public data?
- What does the same method look like in a registry-based market such as Colombia?
- Where do commercial contact databases fall short?
- How do you map the invisible market, step by step?
- How does a market map become a verifiable pipeline?
- What else does this guide cover?
- Frequently asked questions
- Sources
What is B2B market intelligence, and why should a mid-market manufacturer care?
B2B market intelligence is a verified count and description of every company that could buy from you: who they are, where they operate, how they are classified and whether they can be reached. It gives a manufacturer or distributor a denominator, so pipeline can be judged against the real market and not against a list.
Most commercial teams work from lists they inherited: the people who visited a booth, the accounts a distributor opened, the leads that filled a form, the contacts a database vendor sold. Each list is a sample, and each sample was drawn by a channel with its own bias. None of them can tell you how big the market is, because none of them was built to count it.
Market intelligence reverses the order. It starts with the universe of possible buyers, taken from records that exist whether or not anyone ever contacted those companies, and only then asks which of them your current channels reach. The difference between the two is the invisible market: buyers that exist, fit your offer and appear in none of your lists.
Why do trade shows, distributors and inbound leave buyers unseen?
Every channel shows you the buyers it can reach and hides the rest. A trade show shows who walked the floor, a distributor shows its own accounts, inbound shows who searched for you, and a database shows who was already tagged. Together they describe a slice of the market, not its edges.
This is not an argument against those channels. A show remains one of the fastest ways to meet a buyer in person; it was simply never designed to answer how many buyers exist in your region and sector.
“The show introduces the buyers who walked by your booth. We map the ones who didn’t.”
Nicolás Calderón, Founder, DealTunnel
The practical consequence is that a team can do every channel well and still be unable to say what share of its market it has seen. When the pipeline stalls, the cause is usually not effort but coverage: the next buyer was never on the list. Mapping the market first turns that guess into a number you can check.
- Trade shows: bounded by attendance, venue and the buyers who chose to travel.
- Distributors: bounded by the accounts the distributor already serves and chooses to share.
- Inbound: bounded by who already knows to search for you.
- Contact databases: bounded by what the vendor collected, tagged and kept current.
Which public records show the buyers your channels missed?
Four families of public record do most of the work: business registries, labor statistics such as the BLS QCEW, census business statistics such as County Business Patterns, and procurement records. Each is built for administration, not marketing, so each counts companies that never raised a hand.
Business registries. Registries record who exists, under what legal form, with what declared activity, and whether the registration is current. In Colombia that is RUES, the unified commercial registry fed by the chambers of commerce. A registry gives you identity and status, which is what you need before spending anything on outreach.
Labor statistics. The Bureau of Labor Statistics’ Quarterly Census of Employment and Wages (QCEW) is built from state unemployment insurance filings. BLS describes it as “covering more than 95 percent of U.S. jobs available at the county, state, and national levels” (BLS, QCEW overview, page last modified January 28, 2026). It publishes establishment counts by industry, so it works as a census of locations that employ people.
Census business statistics. The Census Bureau’s County Business Patterns “provides annual statistics for establishments with paid employees” at detailed geography and industry level (Census Bureau, About County Business Patterns, retrieved October 13, 2026). It adds employment size classes, which helps you separate a mid-market buyer from a micro-business.
Procurement records. Public tenders and awards show who buys what, in which volume and from whom. They are the strongest evidence of purchasing behavior where they exist, and they are partial: they cover public buyers only. Use them to confirm demand patterns, not to count the whole market.
How do you size your real addressable market from public data?
Pick the industry that buys from you, not the one you belong to, translate it to classification codes, and count establishments in your geography from a public series. The result is a denominator with a source and a date, which you can then narrow by size, location and fit.
The table below is a worked example for Florida. The three figures are different buyer universes depending on what you sell: a building-materials distributor sells to the construction sector (NAICS 23), a manufacturer that sells through distributors sells to wholesalers (NAICS 42), and an industrial-equipment or MRO supplier sells to factories (NAICS 31-33). Each count belongs to the buyer, not to the seller, and it should be read only against that seller’s market.
| Buyer sector | NAICS | Establishments | Share of all private |
|---|---|---|---|
| Construction | 23 | 88,586 | 10.0% |
| Wholesale trade | 42 | 42,050 | 4.7% |
| Manufacturing | 31-33 | 28,099 | 3.2% |
| All private industries (denominator) | 10 | 888,276 | 100% |
Read the table as a method, not as a verdict on Florida. QCEW counts only establishments covered by state unemployment insurance, so sole proprietors without employees are outside it, and an establishment appears under the industry code it reported, which can differ from how the company describes itself. Those limits are the reason the next steps verify each candidate before anyone is contacted.
These numbers are a ceiling, not a target. Establishments include very small sites and branches of the same firm. The next cuts are the ones that make the number useful: county, employment size class from County Business Patterns, and the sub-industry codes at the four- to six-digit level that match what your product actually serves.
What does the same method look like in a registry-based market such as Colombia?
Where labor statistics are less granular, the business registry carries the weight. In Colombia, RUES lets you count registrations by activity code (CIIU), legal form, status and last renewal year, which turns a registry into a market map for any sector.
The table counts registrations for CIIU division 10, food product manufacturing, under one explicit perimeter. These are registrations in a registry, not plants: one business can hold several, and a registration says nothing about production capacity. The count is useful as the upper bound of the set you need to examine, and it is only as current as the extract it came from.
| Scope | Registrations |
|---|---|
| Bogotá chamber of commerce | 4,265 |
| Rest of the country | 8,448 |
| Total CIIU 10 | 12,713 |
| All activities in the same perimeter (denominator) | 847,890 |
The filter choices matter more than the headline number. Dropping the renewal-year filter or adding natural persons changes the count materially, which is why a figure without its perimeter is not a figure. Write the perimeter next to the number every time.
Where do commercial contact databases fall short?
Contact databases are good at one job: giving you a named person at a company you have already chosen. They fall short at the earlier job of defining the market, because their coverage is whatever the vendor collected, and you cannot see the part they missed.
The comparison below is about what each source is built to answer, not about which is better. A serious program uses all three, each for the question it can answer.
| Question | Contact database | Public records | Trade show list |
|---|---|---|---|
| Built for | Reaching a named person | Administration and statistics | Meeting people who attended |
| Who is included | What the vendor collected and tagged | Every registered or covered establishment | Visitors and exhibitors who showed up |
| Can it size the market? | No: no way to see what is missing | Yes: counts with a source and a period | No: attendance, not population |
| Verifiable by you | Only by sampling | Yes: the source is public | Only for those who attended |
| Weak point | Coverage is invisible | No direct contact details | Bounded by one event |
Public records have the mirror-image weakness: they identify the company and its status, and they rarely give you a reachable person. That is why the two are complements. Records define and verify the market; contact data is added afterwards, only for the companies that survived verification.
How do you map the invisible market, step by step?
Seven steps take you from a vague segment to a verified list: define the buyer, translate it to codes, count the universe, cut it, subtract what you already see, verify identity and status, and hand the result to the pipeline with its hypotheses written down.
- Define who buys, not who you are. Write the buyer’s industry and the job your product does for them. A distributor of fasteners is not selling to fastener distributors.
- Translate it to classification codes. Use NAICS in the United States and CIIU in Colombia. Pick the narrowest code that still matches the buyer, and note the codes you excluded and why.
- Count the universe. Pull establishments or registrations for those codes from a public series, with period and source. This is your denominator.
- Cut it. Apply geography, size class and legal form. Record each filter next to the resulting number.
- Subtract what your channels already show. Match your CRM, show lists and distributor accounts against the universe, with exact-match logic, not fuzzy name matching. What remains is the invisible market.
- Verify identity and status. Confirm the registration is active and recent, and that the company is what the code says it is. Discard what fails; do not pay to reach it.
- Write the hypothesis. For each segment, state why you expect it to buy and what would falsify that. Then add contact data only for the verified set.
How does a market map become a verifiable pipeline?
A map becomes a pipeline when every stage is a fraction of the stage before it: universe, verified set, contacted, replied, met, closed. With denominators at each step, a stall points to the stage that failed, instead of to the whole program.
This is the discipline behind the 16 steps DealTunnel runs, from the market map to the signed contract. The map is the first module, and everything downstream is measured against it. Illustratively, if 300 verified companies are contacted and 60 reply, the reply rate is 60 of 300, and you can judge it against the universe you defined.
Because the universe came from records you can check, the pipeline is auditable by someone who was not in the room: your CFO can open the source, read the perimeter and recompute the count. That is what verifiable means here. It does not guarantee a result; it guarantees that the result, good or bad, can be explained.
The method also has limits worth stating. Public records lag: a registry can show a company as active months after it stopped trading, and a quarterly series describes the last quarter, not today. Codes are coarse, so two companies with the same code may sell to very different buyers. And a verified company is not a willing buyer. The map narrows the search and removes guesswork about coverage; it does not replace the conversation that tests whether the offer fits.
For the full model of how the map connects to contact, sales sessions and closing, read the parent guide on revenue engineering. For companies looking at the same method from Latin America, see B2B expansion in Latin America.
What else does this guide cover?
Five deeper articles are planned on this topic. Each one takes a single section of this guide and develops it with its own data. None is published yet; this list shows what is coming and when.
- B1 · December 2026
B2B market intelligence: a complete guide to mapping the buyers your channels never showed you. - B2 · January 2027
The buyers who never walked by your booth: what a trade show cannot count. - B3 · February 2027
Public records for B2B prospecting: registries, labor statistics and procurement. - B4 · March 2027
Size your real addressable market with public data. - B5 · April 2027
Where B2B contact databases fall short in the mid-market.
Frequently asked questions
What is B2B market intelligence?
B2B market intelligence is the practice of building a verified picture of the companies that could buy from you, who they are, where they are and how they are classified, before anyone sends a message. For a mid-market manufacturer or distributor it relies on public records, such as business registries, labor statistics and procurement data, rather than on a purchased contact list alone.
Is a contact database enough to size my market?
No. A contact database lists people someone has already collected and tagged. It does not tell you how many buyers exist in your sector and region, so it cannot give you a denominator. Public counts of establishments or registered businesses give you that denominator, and the database can then be checked against it.
What public data can I use to count potential buyers in the United States?
The Bureau of Labor Statistics QCEW publishes establishment counts by state, county and NAICS industry each quarter, and the Census Bureau's County Business Patterns publishes annual establishment statistics by industry and employment size. Both count establishments, which are locations, not companies.
Do these counts include companies that are not yet customers of my competitors?
Yes, and that is the point. Counts come from administrative records, not from anyone's customer file, so they include buyers that no trade show list, distributor account or inbound form has ever shown you. They do not tell you which of those buyers will respond, which is why the pipeline step tests that separately.
How is this different from outbound prospecting?
Outbound starts from a list and asks who will answer. Market intelligence starts from the full set of possible buyers, sizes it, verifies identity and contactability, and only then decides how to reach them. Prospecting is one use of the map, not a substitute for it.
How long does a first map take?
DealTunnel delivers a Revenue Audit, which includes this market mapping for your sector, within 72 hours. The audit is a first map, not a guarantee of any commercial outcome.
Sources
- U.S. Bureau of Labor Statistics, QCEW overview. “covering more than 95 percent of U.S. jobs available at the county, state, and national levels.” Page last modified January 28, 2026. bls.gov/cew/overview.htm
- U.S. Bureau of Labor Statistics, QCEW Florida area file, 2026 Q1 (area 12000, own code 5, qtrly_estabs: NAICS 23 = 88,586; NAICS 42 = 42,050; NAICS 31-33 = 28,099; all industries, industry code 10 = 888,276). Retrieved October 13, 2026. data.bls.gov/cew/data/api/2026/1/area/12000.csv
- U.S. Census Bureau, About County Business Patterns. “CBP provides annual statistics for establishments with paid employees.” Retrieved October 13, 2026. census.gov/programs-surveys/cbp/about.html
- RUES, Registro Único Empresarial y Social (Colombia). Counts in this guide are DealTunnel’s own query of an extract loaded July 23, 2026 (registrations through July 3, 2026), run October 8, 2026. Registry: rues.org.co
Author: Nicolás Calderón, founder of DealTunnel. Last updated: October 13, 2026.
See the buyers your channels never showed you.
Your first map arrives in 72 hours, in a private hub, with every count tied to its source.