Turn County Business Patterns Into a B2B Target List

If you sell to other businesses, the hardest question in your plan is how many real buyers exist in the area you cover. Most B2B teams answer it with a guess and a round number. The Census Bureau already answered it for you, for free, and updates the answer every year. County Business Patterns counts the businesses in every county, sorted by industry and by how many people they employ. Read it right and it turns a vague market into a countable list of accounts and a territory plan a sales team can actually work.

County Business Patterns, usually shortened to CBP, is an annual release from the Census Bureau. It counts establishments, employment, and payroll for a reference week in mid-March, and it slices those counts by industry code and by employee-size band, all the way down to the county. That is the exact shape a B2B seller needs. You are not trying to size households the way a home-services trade would. You are trying to size firms, and firms are what CBP counts.

This is one of the public datasets we open at the start of every engagement, and it sits inside the five-phase system we run before anything gets built. The same tables the expensive market-sizing decks quietly resell are free at the source. Here is the method to pull them and turn them into a target list.

#Establishments versus employment, and why the difference decides your list

CBP reports two numbers that look similar and mean very different things. Establishments is a count of physical business locations. Employment is a count of the people working in them. If you sell software seats, seats scale with people, so employment is closer to your account value. If you sell a service to a location, a delivery route, a facility contract, or an on-site install, then establishments is your account count, because each site is a potential account.

Get this straight before anything else, because a firm with several locations shows up as several establishments. A regional distributor with four warehouses is one company but four establishments in CBP. Whether that is one account or four accounts depends on how you sell. Decide it up front, then read the column that matches. CBP hands you both, so you never have to pick blind.

• Selling per site, a route, a facility, an install, a per-location subscription? Count establishments. Each location is an account.
• Selling per seat or per headcount? Weight by the employment column, or at least by size band, because a 5-person shop and a 200-person plant are not the same deal.
• Selling to the parent company, not the location? Count establishments to size the market, then expect your real account list to be smaller once multi-site firms collapse into one logo.

#Speak the market's language with NAICS first

CBP is organized entirely by the North American Industry Classification System, NAICS, so you cannot pull it well until you know your buyers' codes. NAICS is a hierarchy. Two digits is a broad sector, three and four narrow it, and six digits is a specific line of business. If you sell CNC tooling to metal shops, you do not want the whole manufacturing sector. You want the four- to six-digit codes that describe metal fabrication, such as the machine-shop and fabricated-metal families under sector 33.

1. Open the official NAICS reference and search the plain-language name of the businesses you sell to. Note every code that fits, at the most specific level that still captures your buyer.
2. Write down two or three codes, not one. Buyers rarely sit in a single tidy code, and CBP lets you pull several and add them up.
3. Decide your digit depth. Six-digit codes give the cleanest buyer match but thinner counts in small counties. Four-digit codes are broader but more stable. Start at four, drill to six where the count is large enough to trust.
4. Keep the list. This same NAICS set feeds every other public-data pull you do, so it is worth getting right once.

#Pull CBP for your counties in under an hour

The front door is data.census.gov, the same site that serves the American Community Survey. You do not need an account or code to start. The CBP program page at census.gov/cbp documents the tables and the size-band definitions if you want the reference alongside. Work through it in this order:

1. Search a CBP table by industry and geography, for example your NAICS code plus your county name. data.census.gov returns the County Business Patterns table with the newest release already selected.
2. Set the geography to the counties you actually cover, not a state. Add each county in your territory so the table sums your real footprint instead of a preset region.
3. Filter to your NAICS codes. Pull each code you wrote down. If you chose several, pull them together so the counts add cleanly.
4. Turn on the establishment-size-class detail. CBP breaks every industry into standard employee-size bands, so you see how many firms have 1 to 4 employees, 5 to 9, 10 to 19, 20 to 49, 50 to 99, and up.
5. Export to CSV. Now you have establishments by NAICS, by county, and by size band, which is the raw material for the whole plan.

As an illustration of the shape, say CBP shows 240 establishments in a fabricated-metal NAICS across your three target counties, and the size bands split roughly into a long tail of tiny shops plus a middle of firms in the 20-to-99-employee range. The exact split is whatever your pull returns. The point is that you now have a defensible count instead of the round number a brief assumed.

#Filter to the size band you actually sell to

This is the step that turns a total count into a target list. A total establishment count almost always overstates your addressable market, because a big share of any industry is one- and two-person firms that cannot buy what you sell. The size bands let you cut them. If your product needs a buyer with an operations team, a budget line, and enough volume to justify the deal, you probably sell to firms in the 20-to-99 or 100-plus bands, not the sole proprietors.

• Set a floor band. Decide the smallest firm that can realistically buy, then drop every band below it. The remaining count is your qualified establishment universe.
• Set a ceiling if you have one. Very large firms often buy through procurement or national contracts you cannot reach as a local seller, so they may belong to a different motion, not your list.
• Sum the qualified bands per county. That per-county number is what you hand a sales team, and it is small enough to be honest about.
• Note the excluded tail. The tiny-firm bands you cut are not worthless. They may be a self-serve or a lighter-touch offer later, so record the count rather than deleting it.

Do the same math the service-area census read does for households, but on firms. Take the qualified establishment count, apply a defensible win rate and an average deal size, and you have a revenue ceiling for the territory that is grounded in a real count, not a hope. If that ceiling is too small to fund the effort, you learned it from a free table instead of a wasted year.

#Split the count into a territory plan

A qualified count by county is already a territory plan in disguise. The counties with the most qualified establishments are where a rep or a campaign earns the most per hour, so they get first coverage. The thin counties get a lighter touch or get grouped into a single blended territory. You are allocating scarce sales and marketing attention against a real distribution of buyers, which is the whole point of doing the count.

The plan drives three concrete decisions. First, where the effort goes: the dense counties get the local SEO and paid budget and a named rep, and the sparse ones get pooled. Second, what you say: an area thick with mid-size manufacturers wants a different message than one full of small professional-services offices, and CBP tells you which you are looking at. Third, whether a segment is worth entering at all: if the qualified count in a NAICS is tiny across your whole footprint, the honest move may be to drop it, and it is far cheaper to learn that from a table than from a quarter of missed quota. Pair the count with a competitor analysis to see who already covers the dense counties before you commit a rep there.

#Refresh it every year, and pair it with the neighbors

CBP publishes annually, so a single pull is a snapshot and a sequence of pulls is a trend. Line up the same NAICS and counties across the two most recent releases and watch the qualified count move. A county where establishments in your target band are rising is a market forming faster than the competition for it, which is the cheapest kind to win. A county where the count is shrinking is a warning to pull effort back before you overinvest. The direction is often worth more than the level, and it is free to read.

CBP is strongest when it is not alone. It tells you how many buyer firms exist and how big they are, but not where their workers cluster or how demand is shifting week to week. Pair it with the BLS and LEHD workforce guide to see where the employment actually sits, and with seasonality reads for timing. CBP is the census of firms; the others tell you how those firms move. Together they turn a static list into a live plan.

That standard is a workflow, not a slogan, and the Census Bureau hands you the hardest part for free. Pick your NAICS codes, draw your real counties, pull the size bands, cut to the firms that can buy, and let the count set the territory. When you are ready to turn the list into a site and a channel plan that reaches those firms, see who we serve for how the work is priced by your industry and your stage.