Feasibility study first
Every engagement starts with a feasibility study. We check what the public record can confirm for your criteria, state by state, before we build anything.
Revenue Estimation Guide
How to estimate a private company's revenue from public signals
Private companies don't publish revenue. Their licenses, filings, job ads, vehicles and permits still show how big they are. This guide shows how to read those signals together and turn them into a range with a confidence level.
By Zan Pogac, Nodior. Published October 18, 2026.
The short answer
To estimate a private company's revenue, collect several independent public signals of its size: headcount from job ads, staff pages and benefit-plan filings, plus licenses, permits, fleet and locations. Multiply headcount by a revenue-per-employee band typical for the industry, check it against a second method such as payroll, and report a range with a confidence level.
Why a private company has no public revenue number
Public companies report revenue every quarter. Private companies in the US have no such duty. State annual reports name officers and a registered agent, rarely revenue, and tax filings stay confidential.
The figures you find online come from somewhere else. Some are modeled from headcount by a data vendor. Others are self-reported in an award list, a trade ranking or a press release. Many are years old, and few say which kind they are. Treat any single published figure as one more signal to test.
The gap is widest in fragmented service markets such as HVAC, plumbing and pest control, where most targets are owner-led and publish nothing about their finances.
Why one signal can mislead
Every public signal has a blind spot. Field technicians rarely keep professional profiles, so a 60-person contractor can look like a 15-person one. A small residential plumber can have more reviews than a commercial contractor several times its size.
Any one of these can be far off on its own. Read together, the errors tend to show, because they rarely all point the same wrong way. One data point rarely says much. When four or five line up, the read holds.
The public signals that matter
These kinds of public data carry real information about a private company's size. Which ones you can use depends on the industry and the state.
| Signal | What it tells you | Watch out for |
|---|---|---|
| Payroll and benefit-plan filings | How many people are on the payroll or in a retirement or health plan. The closest public proxy for headcount. | Not every company files. Counts can include former staff and lag by a year or more. |
| Job ads | Roles, pay, team-size claims and whether the company is adding people. | Constant hiring can mean high turnover. |
| Staff counts on company sites and professional profiles | Named managers, office roles and team depth. | Field staff rarely keep profiles, so trades look smaller than they are. |
| Licenses and license classes | That the company is active and what work it may do. In some states the class caps job size or yearly volume. | A license shows what a company may do and little about how much it does. |
| Permits and public contracts | Work volume and job size over time. | Much service and repair work needs no permit. |
| Fleet registrations and vehicles | Service vehicles, a close proxy for field crews in route-based trades. | Leased, personal and related-company vehicles won't show. |
| Locations and service area | Branches, counties served, yard and warehouse space. | A second address may be a storage unit or a mailbox. |
| Review volume and recency | That the company is active now, and how customer-facing volume moves. | Commercial customers rarely post reviews. |
| Certifications | Trade and manufacturer credentials. Some dealer tiers are tied to purchase volume. | Many belong to one technician and leave with that person. |
The first three measure people. Licenses, permits, contracts, fleet and locations measure work and capacity. Reviews and certifications support the picture but rarely size a company alone.
How signals become a revenue range
To turn signals into dollars, use two methods that rely on different inputs, and let the range cover both.
- Build a headcount. Use at least two sources that don't depend on each other. A benefit-plan filing and a job ad are independent. A staff page and a profile copied from it count once. Decide whether regular subcontractors are in the count.
- Apply a revenue-per-employee band. Multiply headcount by a low and a high figure typical for the industry and service mix. Install work carries equipment through revenue, so it runs higher per head than maintenance work.
- Run a second method. For a payroll-based check, estimate payroll from headcount and public wage data for the trade and region. Then divide by the share of revenue that payroll usually takes in that kind of business. Permit volume or fleet size can add a further check.
- Set the range. Run it from the lower of the two low ends to the higher of the two high ends. If the methods don't overlap at all, the headcount or a band is wrong. Find out which before you report anything.
- Assign a confidence level based on how many independent signals agree.
Revenue is always a range, and these estimates are not audited financials. They tell you which companies are likely the right size, so you can rank targets before anyone calls an owner.
Confidence levels
The confidence level tells you how much evidence stands behind a range.
| Level | What it takes | How to use it |
|---|---|---|
| High | Several independent signals agree, and the two methods give ranges that overlap. | Rank and filter targets on it. |
| Medium | Two signals agree, or one strong signal carries the estimate, such as a recent benefit-plan filing. | Rank on it, and confirm size with the owner early. |
| Low | Signals are thin or point in different directions. | Treat it as a starting point, and ask the owner before you rely on it. |
A low-confidence range still has value. It shows you where the evidence runs out.
Worked example: a commercial HVAC company
Illustrative, not a real company. The company, signals and bands are made up to show the method. Real bands vary by region and service mix.
The example is a commercial HVAC contractor serving two counties in one metro area, with service contracts, repairs and equipment replacement.
The signals
- License: a state mechanical contractor license at the top class, with no cap on job size, held for about 15 years.
- Job ads and staff pages: job ads from the past year say "join our team of 40 technicians." The company site lists a service manager, two project managers and an office team of seven. Total staff is near 50.
- Benefit-plan filing: the latest retirement-plan filing reports 52 participants.
- Fleet: about 30 service vans registered to the company, a fair number for 40 technicians when some work in pairs.
- Permits: about 250 mechanical permits a year in its two counties, steady for three years, mostly commercial replacements.
The signal that disagrees
Only 14 people list the company as their employer on a professional network. Taken alone, that points to about $2 million in revenue. The cause is known: field technicians rarely keep such profiles. The benefit-plan filing and the job ads are independent and agree, so the profile count is set aside and the gap is noted next to the estimate.
Two methods
Headcount times revenue per employee. About 50 employees times $130,000 to $180,000 per employee gives $6.5 million to $9.0 million.
Payroll. Public wage data for HVAC technicians in the region, plus office and management pay, puts total payroll near $3.0 million. If payroll takes 35% to 50% of revenue in this kind of business, revenue falls between $6.0 million and $8.6 million.
The fleet gives a rough third check: 30 vans doing $200,000 to $300,000 of work each comes to $6.0 million to $9.0 million. Steady permit volume says the business is neither growing fast nor shrinking, so today's headcount is a fair base.
The result
Estimated revenue: $6 million to $9 million, high confidence. The low end comes from the payroll method and the high end from the headcount method. Four independent signals agree on about 50 people with a fleet to match, the methods overlap, and the one signal that disagrees has a known cause.
Common mistakes
- Trusting one published number. A revenue figure in a database or a credit file may be modeled, self-reported or years old, and it rarely says which.
- Using the wrong band. A revenue-per-employee figure from residential service work will understate a commercial installer. A figure from install-heavy work will overstate a maintenance shop.
- Missing people who aren't on payroll. Owner-led companies often rely on subcontractors and family members. Leaving them out makes a headcount-based estimate come in low.
- Ignoring dates. A three-year-old filing or a job ad posted before a branch closed describes a different company.
- Mixing related companies. One owner may hold licenses, vehicles and permits under two or three entities. Size only one and you understate the business. Add a sister company that isn't for sale and you overstate it.
When to hand it off
For a handful of companies, you can do this yourself with the steps above. None of the data is secret. It takes time to find, date and cross-check.
The work changes when you need every company that fits your criteria across a state or a region. That can mean hundreds of companies, each sized with the same bands and rules so the ranking holds. Most of the effort goes into collecting and checking. That is the work we take on, and it's part of every acquisition target database we build.
Revenue sizing in a Nodior target database
We build one database for one acquirer: every company we can find in public sources that fits your criteria, with a revenue range and a confidence level for each. See how it works or look at a sample target list.
Ranges you can rank on
Each company gets a revenue range, a confidence level and the signals behind it, next to the owner and whether it is independent or already acquired. You get only the fields you asked for.
Research only
We don't contact owners and take no success fee. Your criteria stay private, with an NDA on request, and we don't resell lists. If a company misses the agreed criteria, the rework is included.
Common questions
How accurate is a private company revenue estimate?
It depends on how many independent signals agree, which is why each estimate carries a confidence level. A low-confidence range is a starting point for your own questions to the owner. None of these estimates are audited financials.
Why give a range instead of a single number?
A single number suggests a precision the public record can't support. A range shows what the evidence supports, and the confidence level shows how much evidence there is.
Is revenue per employee enough on its own?
It's a good starting point, but public headcounts are often wrong for field-heavy companies, and the right band depends on the service mix. Check it against a second method, such as payroll, permit volume or fleet size.
Can you estimate revenue for a company with almost no online presence?
Often, yes. Licenses, permits, vehicle registrations and benefit-plan filings exist whether or not a company markets itself. When the record is thin, we mark the estimate low confidence instead of guessing.
Do you contact the company or its owner to confirm revenue?
No. We only do research from public sources and never contact owners. Your criteria stay private, and we sign an NDA on request.
Scope your research
Send the buyer, vertical, geography and acquisition criteria, and we will tell you whether a target database is a fit. We reply within one business day.