Using US Census and BLS data to segment B2B accounts by region

What to take away

Why Census and BLS data beat vendor firmographics for segmentation

Vendor firmographics arrive as a black box. A record says 240 employees, and you cannot see whether that figure came from a filing, a phone survey or a model. Public data shows its work. The U.S. Census Bureau publishes establishment counts and payroll totals with methodology notes attached.

Coverage is the second difference. Vendors concentrate on accounts large enough to sell to. County Business Patterns covers nearly the whole employer universe, down to small firms in rural counties. That matters when your territory includes secondary metros that vendor files thin out.

Cost is the third. Census and BLS datasets are free to download. A team can spend the saved budget on marketing analytics works instead of another seat license.

Freshness runs the other way. Vendor files update continuously; public data updates on a schedule. Section 7 covers how to work with that.

None of this makes vendor data useless. It makes public data the right base layer, with vendor enrichment on top.

What each source is good at

County Business Patterns: establishment counts by county and industry

County Business Patterns counts establishments, not companies. One firm with four locations appears as four establishments. For territory planning that is usually what you want, because each location can be a buying unit.

The data breaks down by NAICS industry code and by county. You can ask how many establishments in a six-digit NAICS code sit in each county of a state, then roll those counties into territories.

Small-county suppression is the main caveat. Census withholds cells that could identify a single employer. Suppressed cells still count in higher-level totals, so state and national figures stay complete.

A worked example for one territory

Suppose you sell workforce scheduling software to outpatient care centers, NAICS 621498. Pull establishment counts for that code across the counties in your territory. Rank counties by count. A county with 40 establishments is a different prospect than one with 4.

Set a floor: counties below 10 establishments go to a self-serve motion. Counties above 40 get named account treatment. Everything between gets marketing coverage only. That is b2b demand generation metrics made concrete.

Quarterly Census of Employment and Wages for wage and employment filters

QCEW reports employment and wages by industry at the county, metro, state and national level. It comes from unemployment insurance filings, so it covers roughly 95 percent of U.S. jobs. The Quarterly Census of Employment and Wages : U.S. Bureau of Labor Statistics publishes these figures quarterly.

Employment counts tell you how many people work at establishments in an industry. Average weekly wage tells you what they earn. Together they separate a 500-person staffing agency from a 500-person software firm.

Wage filters matter for products priced per seat. A county with high average wages in your target industry can support higher contract values. A county with the same establishment count but low wages may need a lighter package.

Building a wage filter

The floor is a judgment call. Test it against closed-won accounts before locking it in.

A typical starting floor is an average weekly wage of $1,200. Pull the QCEW series for your target NAICS codes, sort counties by that figure, then check which side of $1,200 your closed-won accounts sit on.

Adjust the number until the filter matches reality. Sectors with high capital intensity usually carry a higher floor. Retail and food service accounts usually carry a lower one.

Two thresholds beat one. Counties above the floor get a full package. Counties below it go to a self-serve or digital motion, and the middle band gets marketing coverage.

BLS industry and occupation data for buying-committee sizing

Occupation data answers a different question: who inside the account will join the buying committee. The Overview of BLS Statistics by Occupation : U.S. Bureau of Labor Statistics lists employment by occupation, including detailed SOC codes.

If you sell to finance teams, find the count of accountants and auditors in your target metros. That count, not total headcount, is the addressable population. A 2,000-person logistics firm may have fewer finance seats than a 400-person accounting firm.

Use occupation ratios to size committees. If your product touches three roles, multiply the relevant occupation counts by the ratio of workers likely to sit on a committee. Ratios are estimates, so treat them as ranges.

Mapping roles to filters

Each role becomes a filter. Accounts missing the champion code drop out of the named list.

Start with a role map. Finance leader: SOC 11-3031, financial managers. Practitioner: SOC 13-2011, accountants and auditors. Operations lead: SOC 11-1021, general and operations managers.

Committee sizing needs a multiplier. A common starting ratio is 0.1, one committee seat per ten workers in the mapped occupation code. Treat that ratio as a starting estimate and tune it against closed-won data.

For example, a 400-person accounting firm with 60 accountants and auditors yields about six likely committee members at a 0.1 ratio. The filter uses that number, not total headcount.

Geographic BLS data for regional account tiers

The Overview of BLS Statistics by Geography : U.S. Bureau of Labor Statistics organizes data by census region, state and metro area. That structure maps cleanly onto sales territories.

Metro-level employment trends show which regions are adding jobs in your target industries. A metro adding 5,000 jobs in a sector over a year is a growth market. A flat metro is a replacement market.

State policy differences also shape demand. California, Texas and New York run large regulated sectors with their own employment rules. Massachusetts, Illinois and Washington do the same in education, finance and technology. Pair the BLS geography data with state labor department releases when a sector is policy-sensitive.

Assigning territories

Start with the Databases, Tables & Calculators by Subject : U.S. Bureau of Labor Statistics to pull series for each metro in your plan. Group metros into territories so each rep gets a similar mix of tier one and tier two accounts. Balance by opportunity, not by geography alone.

Turning datasets into account selection filters

A filter is a rule that decides whether an account enters a named list. Public data gives you four filter families: establishment count, employment size, average wage and occupation mix. Combine them into a score.

The output is a ranked territory map. It is not a contact list, and it should not be treated as one.

Filter checklist

This is the same discipline that separates working selection, plays and proof in any account-based program.

Keeping segmentation current with BLS release schedules

QCEW publishes quarterly, with a lag of roughly five to seven months after the quarter ends. County Business Patterns publishes annually, about a year after the reference year. Occupation data updates less often.

If you rebuild filters once a year, your territory plan drifts for months. The fix is a calendar keyed to the Schedule of Selected Releases for September 2026.

A refresh cadence

Step four is where the filter meets reality. If a tier one county produces no pipeline for two quarters, the filter is wrong or the coverage is missing.

Limits of public data and where to supplement it

Public data stops at the establishment. It does not name the company, the buyer or the technology in use. It also lags reality by months, and it suppresses small cells.

Supplement it with three things. First, firmographic enrichment for company names and contacts. Second, intent or technographic signals for timing. Third, your own CRM history for proof that a filter predicts revenue.

Compliance matters when you combine sources. The Federal Trade Commission and state attorneys general enforce rules on data use, and the SEC governs what public companies disclose about their workforce. Keep a record of where each field came from.

The Small Business Administration publishes size standards that define which accounts count as small business for contracting purposes. Those definitions affect eligibility in some sectors, so check them before you build a small-business segment.

Finally, treat the score as a hypothesis. The b2b pipeline velocity calculation your team already tracks is the test. If a filter does not move pipeline, retire it.

Common questions