Business Valuation: Why Founders and Buyers Never Agree on What a Company Is Worth

In the summer of 2006, Terry Semel flew to Palo Alto with an offer that most twenty-two-year-olds would have accepted before the sentence was finished. Semel was the CEO of Yahoo, and he was offering $1 billion in cash for Facebook. Mark Zuckerberg, who had launched the site from a Harvard dorm room just two years earlier, listened to the number and turned it down.

The reaction inside Facebook was not unanimous. Several board members urged Zuckerberg to take the deal. Peter Thiel, one of Facebook's earliest investors, reportedly questioned the refusal. A billion dollars was real money. Facebook had roughly 9 million users, almost no revenue, and a website that still ran on servers in a rented office. The company had never been profitable. By any conventional valuation method, $1 billion was generous.

Zuckerberg said he saw something Yahoo didn't. In a meeting with his board, he said he believed Facebook could be worth "much, much more." He was right. By 2012, Facebook went public at a $104 billion valuation. By 2024, Meta Platforms was worth over $1.5 trillion. The billion-dollar offer now looks like the single best decision Zuckerberg ever made.

But here is the question nobody asks about that meeting: Was Zuckerberg's refusal the product of superior vision, or was his brain doing exactly what every brain does with things it owns?

The answer, based on five decades of behavioral economics research, is almost certainly both. And the tension between those two explanations is the central problem of business valuation. Founders and buyers systematically disagree about value because they are running different cognitive software. The founder's brain inflates the value of what it built. The buyer's brain deflates the value of what it doesn't yet own. Neither is seeing the company clearly. The gap between their estimates isn't a negotiation tactic. It's a neurological artifact.

What Is Business Valuation and Why Is It So Contested?

Business valuation is the process of estimating the economic value of a company. The methods are well-established. Discounted cash flow analysis projects future earnings and discounts them to present value. Comparable company analysis values the business relative to similar companies that have been sold or are publicly traded. Asset-based valuation adds up what the company owns and subtracts what it owes.

Each method is rigorous. Each produces a different number. And the range between them is often vast, not because the methods are flawed but because valuation is, at its core, an exercise in predicting the future, and the future is uncertain.

What the methods cannot account for is the psychological relationship between the valuer and the valued. And that relationship, research consistently shows, is the largest variable in the equation.

Richard Thaler ran the experiment that changed how economists think about ownership in 1990. He and his colleagues gave coffee mugs to half the participants in a room and asked them what they'd accept to sell the mug. The other half, who didn't receive a mug, were asked what they'd pay to buy one. The sellers demanded roughly twice what the buyers were willing to pay. Same mug. Same room. Same moment. The only difference was possession.

Thaler called it the endowment effect. The brain assigns a valuation premium to things it owns, and the premium is not small. Across hundreds of replications, the ratio holds: people value what they have at 1.5 to 3 times what they'd pay to acquire it. The effect persists in experts, in repeated trials, and in real-world transactions with real money at stake.

Now scale the coffee mug to a company. A founder who has spent five years building a business, who has poured in personal savings, relationships, reputation, and years of cognitive and emotional labor, is experiencing the endowment effect at a magnitude that a coffee mug cannot begin to approach. The company isn't just something they own. It's something they built with their hands, which triggers the IKEA effect (people value things they assembled themselves 63 percent higher than identical pre-assembled items, per a 2012 study by Michael Norton, Daniel Mochon, and Dan Ariely). It's also something they've suffered for, which triggers what researchers call the "effort heuristic": the brain uses the difficulty of creating something as a proxy for its value.

The buyer, standing on the other side of the table, has no such attachment. The buyer evaluates the business using the same cognitive machinery they'd use to evaluate any potential acquisition: what will this asset return relative to its cost? The buyer's brain isn't deflating the value out of malice. It simply hasn't built the endowment, the IKEA premium, or the effort heuristic. The company is a spreadsheet, not a legacy.

The gap between these two perspectives is the fundamental friction of every acquisition, every fundraising round, and every exit strategy. It is the reason founders feel insulted by offers that buyers consider generous.

Here's the napkin line: the founder sees the company through the lens of what they put in. The buyer sees it through the lens of what they'll get out. Neither lens is wrong. They're pointed at different things.

Was Zuckerberg Right or Endowed?

The Facebook-Yahoo negotiation is usually told as a story about visionary founders and short-sighted corporations. The narrative is satisfying. It's also incomplete.

In 2006, Zuckerberg owned approximately 30 percent of Facebook. A $1 billion acquisition would have given him roughly $300 million at age twenty-two, more wealth than he could spend in several lifetimes. The rational calculation, weighted by the probability of Facebook actually becoming worth $100 billion versus the certainty of $300 million, was not obviously in favor of refusal. The expected value calculation depends on assumptions about probability that no one, including Zuckerberg, could have verified at the time.

What Zuckerberg had, in addition to whatever strategic insight informed his decision, was a brain running every cognitive bias that inflates the perceived value of what you've built. He'd created Facebook. He'd coded significant portions of it himself. He'd turned down a previous $75 million offer from Viacom, which means his reference point had already been anchored upward (a phenomenon documented in anchoring bias research). Each previous offer refused created a commitment that, per Cialdini's consistency principle, made refusing the next one psychologically easier.

None of this means Zuckerberg was wrong. It means the cognitive forces that would have driven him to refuse even a terrible offer were identical to the ones that drove him to refuse an offer that happened to be too low. The endowment effect doesn't care whether you're right. It inflates your valuation either way.

Daniel Kahneman, asked about this dynamic in business negotiations, put it directly: "The seller's reluctance to part with the asset is not a signal of the asset's value. It's a signal of the seller's attachment."

The practical consequence for founders approaching any valuation event, whether a fundraise, an acquisition offer, or an internal assessment, is that the number in their head is contaminated by endowment before the calculation begins. This doesn't mean the number is wrong. It means the founder cannot distinguish between "I believe this is worth more because I see something others don't" and "I believe this is worth more because my brain systematically overvalues things I own." The phenomenological experience of both is identical. They feel exactly the same.

The Sunk Cost Shadow in Every Valuation

There is a second bias that compounds the endowment effect in business valuation, and it operates so quietly that most founders never detect it.

Sunk costs are expenditures that cannot be recovered. Time, money, and effort already spent. Economic theory is clear: sunk costs should be irrelevant to forward-looking valuations. A business is worth what it will generate in the future, not what it cost to build in the past. The market doesn't care that you spent three years and $500,000 developing the product. It cares whether the product will generate returns.

But the brain cares deeply about sunk costs. Hal Arkes and Catherine Blumer's classic 1985 experiments showed that people continue investing in failing projects specifically because they've already invested, not because the forward-looking economics justify continuation. The sunk cost fallacy is one of the most robust findings in behavioral economics.

In valuation, sunk costs appear as an invisible floor. The founder who invested $2 million building the company has enormous difficulty accepting a valuation below $2 million, even if the forward-looking cash flows justify only $500,000. The $2 million isn't relevant to the valuation. It's a historical fact about the past, not a prediction about the future. But the brain treats it as a minimum acceptable price because selling below cost triggers the same loss-aversion circuitry that makes people hold losing stocks and continue funding doomed projects.

Research by Terrance Odean at UC Berkeley documented this exact pattern in stock markets. Individual investors hold losing stocks significantly longer than winning ones, a phenomenon Odean called the "disposition effect." The driver is loss aversion: selling below the purchase price crystallizes a loss, and the brain finds the psychological pain of that crystallization more aversive than the economic pain of continued decline. Founders exhibit the same pattern with their businesses. Holding on to a company valued below its sunk costs feels like patience. Selling at a loss feels like admitting failure. The brain will rationalize the first and resist the second regardless of the economics.

The buyer, who has no sunk costs in the business, is immune to this bias. The buyer sees only forward-looking value. This asymmetry creates a specific pattern in acquisition negotiations: the founder's floor is anchored to sunk costs, the buyer's ceiling is anchored to projected returns, and the gap between them has nothing to do with disagreement about the business and everything to do with different brains processing different information.

How Does Uncertainty Warp Valuation on Both Sides?

Ellsberg's paradox, demonstrated by Daniel Ellsberg in 1961, reveals a pattern that operates in every valuation negotiation. Ellsberg offered participants a choice between two gambles. Gamble A: draw a ball from an urn containing fifty red and fifty black balls. Win $100 if you draw red. Gamble B: draw from an urn containing one hundred balls in an unknown ratio of red and black. Same payout. Same expected value. Participants overwhelmingly chose Gamble A. They preferred the known risk to the unknown risk, even though the expected value was identical.

This is ambiguity aversion, and it is distinct from risk aversion. People don't just dislike risk. They dislike not knowing the probability distribution of the risk. And business valuation is, at its core, an exercise in ambiguity. The future revenue of a company is not a known probability distribution. It's an estimate built on assumptions, and the number of assumptions compounds with the time horizon.

For buyers, ambiguity aversion produces a systematic discount. The buyer doesn't know what they don't know about the business. Hidden liabilities, customer concentration risk, key-person dependence, competitive threats that aren't visible from outside. Each unknown produces an ambiguity discount that the buyer applies intuitively, often without explicitly quantifying it. The buyer's offer is lower not because they've calculated a lower value but because their brain is penalizing the unknowns.

For founders, ambiguity operates differently. The founder knows the business intimately, which reduces their experienced ambiguity. But this familiarity creates its own bias. Psychologists call it the illusion of knowledge: the more information someone has about a domain, the more confident they become in their predictions, even when the additional information doesn't actually improve predictive accuracy. Philip Tetlock's research on forecasting, published in Expert Political Judgment in 2005, showed that experts with deep domain knowledge were no more accurate than non-experts at predicting future events, but were substantially more confident.

The founder's deep knowledge of the business produces confident forward projections. The buyer's relative ignorance produces cautious ones. Both are responding rationally to their information environment. Neither is seeing the company objectively. And the gap between confident projection and cautious projection is often the gap between a deal that closes and one that doesn't.