Company size data: how to trust a headcount field

November 13, 2026

You're building a target account list, and the headcount field says one company has 47 employees while another has 312. You segment, personalize, and reach out—only to discover the first is actually a 200-person firm and the second is a 12-person startup. Exact headcount numbers are the least reliable data in most B2B databases. Here's how to stop trusting them blindly.

Why Exact Numbers Lie

Headcount is a moving target. Companies hire, lay off, and reclassify contractors constantly, but database updates lag by months. A field that says 250 might reflect last year's snapshot. Worse, many providers estimate headcount by scraping LinkedIn profiles, which undercounts field workers and overcounts sales-heavy firms where everyone lists a title. Some vendors even interpolate missing data from revenue models or industry averages, producing a precise-looking number with no real basis.

The result: two records for the same company can differ by 300%. If you filter a prospecting list by '51-200 employees' and trust the exact field, you'll include 20-person shops and exclude 500-person enterprises. That pollutes your pipeline with poor-fit leads and wastes sales time on accounts that will never convert.

Treat any exact headcount as a guess, not a fact—especially if the source is a single provider.
Check the 'last updated' date on the field; if it's older than six months, discount it heavily.
Watch for suspiciously round numbers (50, 100, 500) that suggest estimation, not measurement.

Ranges And Bands Beat Precision

Instead of chasing exact headcount, use ranges or bands. Most segmentation decisions don't need to know if a company has 187 or 193 employees—they need to know if it's a small business, mid-market, or enterprise. Standard bands like 1-10, 11-50, 51-200, 201-500, 501-1000, and 1001+ are robust to minor headcount fluctuations and data errors. Even better, many databases expose a 'size range' field that's derived from multiple signals, not a single scrape.

When you build segments, filter on the range field rather than the exact number. This reduces false positives from stale data. For example, if you're targeting companies with 100-500 employees for a mid-market sales motion, a range of '51-200' might still be worth including if other signals (revenue, tech stack) align—because the exact number is unreliable anyway. Ranges give you a buffer against noise.

Another practical approach: bucket exact numbers yourself into custom bands and then audit a sample. If 10% of records in your '201-500' bucket turn out to be under 100 when cross-checked, you know your segment needs a wider margin.

Always prefer a range field over an exact headcount field when available.
Define your own bands based on your sales model (e.g., SMB, mid-market, enterprise) and map exact numbers into them.
Audit a random sample of records per band to measure error rates and adjust boundaries.

Cross Check With Public Sources

Before you trust a headcount for segmentation—or worse, for personalization—sanity-check it against public sources. LinkedIn company pages show an employee count range and list of people; if the page says 11-50 but your database says 480, that's a red flag. Crunchbase and PitchBook often list employee counts for funded companies, and the company's own 'About' page or press releases may mention headcount. For public companies, 10-K filings include exact employee numbers (though they may include contractors).

A quick cross-check takes 30 seconds per record but saves hours of wasted outreach. For high-value accounts, go further: search for recent news about hiring sprees or layoffs, check job postings to infer growth, or use tools like Clearbit or ZoomInfo as a second opinion. If two independent sources agree within 20%, you can trust the range. If they conflict wildly, mark the record as 'size uncertain' and either exclude it from size-based segments or verify manually.

Remember that different sources count differently: LinkedIn counts self-reported profiles, the company counts W-2 employees, and databases may count contacts. A 50-person company might have 120 LinkedIn profiles if they include contractors and alumni. Always ask: what is this number actually counting?

Compare your database's headcount to LinkedIn's employee range and Crunchbase—if they diverge by more than 2x, distrust the record.
For public companies, check SEC filings for exact headcount and note the filing date.
Use a second data provider (e.g., Clearbit, ZoomInfo, Apollo) as a tiebreaker for critical accounts.

Segment Without Size Data

What if headcount is missing entirely? Don't let that block your segmentation. You can build effective B2B segments using alternative signals that are often more reliable and more predictive of fit. Revenue range, funding stage, tech stack (e.g., uses Salesforce or HubSpot), industry, and geographic location are all strong proxies for company size. For example, a company using enterprise-grade tools like Marketo and SAP is likely larger than one using Mailchimp and QuickBooks.

Another approach: use firmographic signals like number of job postings, web traffic (via SimilarWeb), or social media follower counts to infer size bands. A company with 50 open roles is probably growing fast and larger than its headcount suggests. You can also segment by 'digital footprint'—number of employees on LinkedIn, number of locations, or presence of a dedicated sales team—which often correlates with size but is more stable than a single headcount field.

If you must have a size proxy, consider using a composite score: average the available signals (LinkedIn range, revenue band, tech stack tier) and assign a confidence level. Then run your lead generation and prospecting efforts on high-confidence records first, and treat low-confidence records as a separate, lower-priority pipeline.

Use revenue range, funding stage, and tech stack as size proxies—they're often more accurate than headcount.
Infer size from job postings, web traffic, or LinkedIn follower count when headcount is missing.
Create a 'size confidence' score and prioritize outreach to records with high confidence.

Sanity Check Before You Segment

Before you apply any size-based filter to a list of contacts, run a quick sanity check on the underlying records. Start by filtering for outliers: companies with 1 employee but a Fortune 500 domain, or 10,000 employees but a Gmail address. These are almost always data errors. Next, check the distribution of headcounts in your list—if you see a spike at exactly 50 or 100, that's a sign of estimation. Finally, cross-reference a sample of 20 records against LinkedIn and note the error rate.

Build a simple validation rule into your workflow: if the exact headcount differs from the LinkedIn range by more than one band, flag the record for review. If you're using a CRM like HubSpot or Salesforce, you can automate this with a formula field or workflow. The goal isn't perfect data—it's knowing which records are trustworthy enough to segment on, and which need a manual look before you reach out.

Flag records where exact headcount is outside the LinkedIn employee range by more than one band.
Watch for round-number spikes (50, 100, 500) that indicate estimation.
Manually verify a random sample of 20 records per segment to estimate error rate.

Headcount is a useful signal, but only when you treat it as a range, not a fact. Cross-check before you segment, use proxies when data is missing, and always sanity-check a sample. Your pipeline will thank you with fewer wasted outreach attempts and better-fit leads.

Useful links

FAQ

How accurate is LinkedIn's employee count?

LinkedIn shows a range based on self-reported profiles, so it undercounts field workers and overcounts sales-heavy firms. It's best used as a sanity check, not a precise number. Treat the range as a directional signal and cross-check with other sources.

What's the best headcount range for B2B segmentation?

Use bands that align with your sales model—commonly 1-10, 11-50, 51-200, 201-500, 501-1000, and 1001+. These bands are wide enough to absorb data noise while still being actionable for lead generation and prospecting.

Should I trust a database that provides exact headcount numbers?

No. Exact numbers are often estimates or stale snapshots. Always check the last-updated date and cross-check with at least one public source. If the number is critical, verify manually before you segment or personalize outreach.

Can I segment by company size without headcount data?

Yes. Use revenue range, funding stage, tech stack, job postings, or web traffic as proxies. These signals are often more reliable and more predictive of fit than a single headcount field.