How Strategies and Revenue Estimation for Private Companies Is Done

Public companies are legally required to disclose revenue every quarter. Private companies are not, and the vast majority of the economy is private. That single fact forces analysts, investors, and competitors into a discipline that never appears on an income statement: reconstructing a number that was never meant to be seen. Epignosis Insights' review of government business data, industry-association benchmarks, and disclosures buried inside public companies' own filings shows that private-company revenue estimation is not guesswork dressed up as science; it is a set of triangulation methods, each with known error bands, that gets more reliable the more independent signals are stacked against each other.

Why the Data Gap Exists in the First Place

The scale of the problem is easy to underestimate. U.S. Census Bureau Nonemployer Statistics counted roughly 30 million business establishments with no paid employees in 2023, up 25 percent from 2015, and nonemployer establishments alone accounted for about 6.4 percent of U.S. GDP, or $1.8 trillion in receipts, that year. None of those businesses file a 10-K, and most of the millions of small-to-midsize employer businesses above them do not either. Revenue estimation exists because disclosure is the exception in the private economy, not the rule, and any methodology has to start from that imbalance rather than treat missing data as an edge case.

Reading Proxy Signals Where Financials Don't Exist

The starting toolkit is proxy-based: headcount ratios against a comparable public company, real estate footprint, hiring velocity on job boards, and, for regulated sectors, licensing or permit filings. The American Investment Council, the U.S. private equity industry's trade association, reports that roughly 85 percent of private equity investment flows into businesses with fewer than 500 employees, meaning the overwhelming majority of PE-backed companies are exactly the kind of business with no public filings and only partial state-level disclosure. Analysts sizing those companies typically anchor to a public comparable's revenue-per-employee ratio, adjust for sector and geography, and treat the result as a range rather than a point estimate, because a single proxy multiplier can be off by a wide margin if the comparable company's business mix does not match cleanly.

How Data Platforms Turned Estimation Into a Product

Estimation has also become a commercial category in its own right. Morningstar, Inc.'s first-quarter 2026 results show its PitchBook segment contributing $172.4 million to consolidated revenue, up 5.3 percent, or 4.8 percent organically, with management specifically citing new daily valuation estimates for venture-backed companies as a growth driver. That business exists precisely because standard disclosure does not cover most of the venture-backed universe: separate PitchBook data cited alongside that same product launch noted that nearly half of private-market unicorns had gone unrevalued for three years or more. When even professional data platforms are re-estimating valuations that infrequently, any revenue figure pulled from a single cached source should be treated as dated until independently refreshed.

Why Dealmakers Still Can't Agree on the Number

Even with better tooling, disagreement over private-company financials remains the single biggest obstacle to closing deals. PwC's Global M&A Outlook 2026 identifies valuation gaps, alongside geopolitical uncertainty and tight financing conditions, as the primary barrier to deal execution this year, and notes that private lenders financed roughly 80 percent of global leveraged buyouts in 2024 and 2025 as sponsors sought bilateral structuring precisely because syndicated markets demand more standardized disclosure than many targets can produce. A revenue estimate that both a buyer and seller will accept is, in practice, negotiated as much as it is calculated.

Reading Between the Lines of Disclosed Run-Rates

The clearest recent illustration of triangulation in action comes from outside traditional finance altogether. Forbes' 2026 AI 50 reporting pieced together that OpenAI's annualized revenue run rate surpassed $25 billion by late February 2026 and that Anthropic's crossed $30 billion by early April, using disclosed run-rate figures, funding-round terms, and investor commentary rather than any audited financial statement, since both companies remain private. That reporting also found the 50 listed companies had collectively raised $305.6 billion, with two companies accounting for roughly 80 percent of that total. It is a useful reminder that even the most closely watched private companies in the world are sized through inference, not disclosure.

Epignosis Insights' Triangulation Method

Given how thin and how commercially contested private-company data remains, Epignosis Insights never relies on one signal. Every private-company revenue estimate is built by cross-referencing at least three independent inputs: a government or industry-association base rate for the sector and firm-size bracket, a proxy multiple drawn from the closest available public comparable, and, where they exist, secondary signals such as hiring velocity, real estate footprint, or disclosed funding terms. Estimates are always delivered as a range with the underlying assumptions stated explicitly, because a private-company revenue figure without a stated method and a stated error band is not an estimate; it is a guess wearing a decimal point.