--- slug: startup-equity-evaluation type: pattern summary: "How a candidate reads a startup equity offer for what it is actually worth — the five questions that turn a headline grant into a probability-weighted number." created: 2026-05-26 updated: 2026-05-26 related: equity-compensation-types: relation: depends-on note: "Valuing a grant starts with knowing the instrument — ISO, NSO, or RSU — because the form determines the tax treatment and the exercise mechanics that a dollar figure hides." dilution: relation: uses note: "A grant's percentage is a snapshot that shrinks at every future round, so reading an offer means modeling the dilution the headline number ignores." vesting-cliff: relation: depends-on note: "The grant is a ceiling earned over time, so the vesting schedule and the cliff decide how much of the offer a departing employee actually keeps." cap-table-hygiene: relation: informed-by note: "The fully-diluted share count and the option-pool size that an honest evaluation needs are exactly the numbers a clean cap table records and a messy one obscures." total-compensation-architecture: relation: contrasts-with note: "This is the candidate reading the same numbers the founder used to build the offer; the employer-side framework prices the grant that this one decodes." acquisition-exit: relation: uses note: "Expected value depends on the exit, and acquisition — the realistic outcome for most venture-backed companies — sets the scenario the candidate should weight most heavily." liquidation-preference: relation: downstream-of note: "An employee's common stock sits behind every preference in the stack, so the preference terms cap what a grant returns before the headline valuation ever reaches it." --- # Startup Equity Evaluation > **Pattern** > > A named solution to a recurring problem. *Reading a startup equity offer for its real, probability-weighted value rather than the headline number, so the cash-versus-equity tradeoff is made on evidence instead of hope.* A recruiter sends an offer: base salary fifteen percent under what a public company would pay, plus "0.5% of the company" in options. The percentage sounds like a stake worth having. It's also nearly meaningless on its own. Half a percent of what? Today's shares, or the fully-diluted count after three more rounds? At what strike price, with what tax bill on exercise, surviving how much dilution, paying out behind how large a preference stack, in which of the exit scenarios that actually happen? The offer letter answers almost none of this, and the gap between the number on the page and the number an employee can expect to realize is where most of the disappointment in startup equity lives. ## Context This decision sits on the talent side of the [talent-equity](talent-equity.md) part of the lifecycle, at the moment a candidate or early employee weighs a startup offer against a market-rate alternative. It applies to anyone trading cash compensation for equity: the first engineer, the early product hire, a senior operator joining pre-Series-A, a candidate choosing between two startups at different stages. The offer is the output of the founder's [total compensation architecture](total-compensation-architecture.md): the same grant, viewed from the other end of the table. The founder priced it against a salary band and an option pool; the candidate has to decode it back into expected value, usually with less information, under more time pressure, and with no framework for what the numbers mean. ## Problem A candidate must convert an equity offer into a single comparable number (the expected value of the grant, net of tax and weighted by the probability of each outcome) at a point when the company that backs it has no liquid market, no guaranteed exit, and a cap table the candidate hasn't seen. The headline figures an offer leads with (a percentage, or a "value" computed at the last round's price) systematically overstate what an employee will realize, because they ignore dilution, preference, vesting risk, exercise cost, and the base rate of startup failure. The result is a transaction priced on optimism: people accept grants they can't value, discover the gap years later, and conclude that startup equity is a lottery, when in fact it's a knowable, if uncertain, expected-value calculation. ## Forces - **Percentage versus dollar value.** A percentage decays with every round of dilution; a dollar "value" assumes a price that may never recur. Each framing flatters the offer differently, and the honest number requires translating between them at the company's likely future share count. - **Upside versus base rate.** The grant's value in a great outcome is real and large, and it's also the outcome least likely to occur. Weighting only the upside ignores that most startups return zero to common stock; weighting only the base rate ignores why anyone takes the risk at all. - **Information asymmetry.** The company knows the fully-diluted share count, the preference stack, and the option-pool size; the candidate often gets a percentage and a valuation and is left to infer the rest. The questions that close the gap are answerable, but the candidate has to know to ask them. - **Negotiating equity costs goodwill at the worst time.** Pressing for the share count and preference terms can read as distrust in the first conversation with a future employer. But signing on a number you can't value is how the resentment surfaces later, when it's harder to fix. - **Cash now versus equity maybe.** The salary cut is certain and immediate; the equity is contingent and years out. The tradeoff isn't abstract; it's rent, runway, and how long the candidate can personally afford to bet. ## Solution **Translate the offer into a probability-weighted expected value by answering five questions, and treat any number the company will not give you as a finding in itself.** The grant's headline framing is the company's most flattering view of it; the candidate's job is to reconstruct the realistic one. The five questions that turn an offer into a number: 1. **What fraction of the company is this, fully diluted?** Not shares, not last-round dollars, but the percentage of the fully-diluted share count, which includes all options, warrants, and unconverted [SAFEs and notes](safe-note.md). A grant quoted in raw share count with no denominator is unanswerable until you have the denominator. 2. **What is the strike price, and what will it cost to exercise?** For options, the strike is what you pay to convert them to shares. A large grant with a high strike and a short post-termination exercise window can be functionally worthless to someone who leaves before a liquidity event and cannot afford to exercise. 3. **How much dilution is ahead?** Each future round issues new shares and shrinks existing percentages. A 1% grant at seed is routinely a fraction of that by exit. Model the rounds the company will plausibly raise; the [dilution](dilution.md) is not a risk to the grant, it is a certainty. 4. **What is ahead of you in the stack?** Common stock pays only after every [liquidation preference](liquidation-preference.md) is satisfied. In a modest exit, a heavy preference stack can route most of the proceeds to investors before employees see a cent, regardless of the valuation the offer cited. 5. **What is the realistic distribution of exits?** Weight the outcomes that happen, not the one in the pitch. [Acquisition](acquisition-exit.md) is the path most venture-backed companies that exit at all actually walk, usually at a fraction of the unicorn number; a meaningful share return nothing to common. The expected value is the sum across outcomes, each multiplied by its probability, not the best case in isolation. A workable shorthand for the calculation: ``` expected value = Σ (exit_proceeds_to_common × your_diluted_% × P(outcome)) − exercise_cost − tax ``` The discipline isn't precision (the inputs are genuinely uncertain) but honesty about the shape. An offer that survives this translation and still beats the cash alternative is a real opportunity. One that only looks good before you run it is the more common case, and recognizing which one you're holding is the entire point. > **⚠️ How offers obscure value** > > Three framings recur. A grant quoted as a **percentage with no fully-diluted denominator** hides how much pre-existing dilution it already sits behind. A grant quoted as a **dollar value** silently multiplies the share count by the last round's price, a number with no guarantee of recurring and every incentive to be high. And a **four-year value** presented as if it vests on day one ignores that you earn it over time and forfeit the unvested remainder if you leave. None of these is necessarily dishonest; all three flatter the offer, and the candidate is the one who has to deflate them. ## How It Plays Out Consider two offers a senior engineer is weighing. The first is from a Series B company: 0.15%, fully diluted, with a strike set at the last 409A valuation, vesting over four years. The second is from a seed-stage company: "1% of the company," quoted as a percentage with no denominator, salary twenty percent lower. The seed offer's 1% looks like nearly seven times the stake. But it's 1% of a company that will, if it succeeds, raise three or four more rounds before any exit, each diluting that 1%; a seed grant landing near 0.2–0.3% by a late-stage exit is unremarkable. It also sits behind whatever preferences those rounds carry, and the company is at the stage where the base rate of failure is highest. The Series B grant is smaller as a percentage but later in the dilution path, behind a known (if larger) preference stack, at a company that's already survived the riskiest years. Which offer has the higher expected value isn't obvious from the headline numbers, and answering it requires the share count and preference terms that only the question forces into the open. The instructive cases are the ones where the answer arrives too late. An employee accepts a generous-sounding option grant, works four years, and learns at acquisition that a 1x participating preference and a modest sale price leave common stock with little after the investors are paid: the valuation was real, but the [liquidation preference](liquidation-preference.md) ahead of the common claimed the proceeds. Another exercises early into a high valuation, owes alternative minimum tax on the paper gain, and then watches the company fold, leaving a real tax bill against shares that never became cash. Neither outcome was hidden; both were answerable from the cap table and the term sheet at the moment of the offer. The pattern isn't bad luck. It's an evaluation that was never run. ## Consequences **Benefits.** A candidate who runs the five questions makes the cash-versus-equity tradeoff on evidence: they can compare two startup offers honestly, weigh a startup offer against a public-company alternative, and decide how much certain salary they're willing to forgo for a contingent stake they've actually sized. The questions also surface the company. A founder who answers the fully-diluted percentage and the preference terms plainly is signaling a clean [cap table](cap-table-hygiene.md) and a culture of candor; one who deflects is signaling the opposite, and that signal is worth as much as the numbers. **Liabilities.** The inputs are uncertain, and a probability-weighted number carries false precision if mistaken for a forecast: the calculation disciplines the decision, it doesn't predict the outcome. Asking the hard questions early can strain a nascent relationship with a future employer, and a candidate has to judge how hard to press against how much they want the role. And the most rigorous evaluation can't rescue anyone from the base rate; most startup equity is worth little, and the analysis simply tells you that before you sign rather than after. The point isn't to find the offer that pays off. It's to take the equity risk with eyes open, having sized the bet rather than guessed at it. ## Sources - [Carta's equity and compensation data](https://carta.com/learn/) — the benchmark source for grant sizes by role and stage, dilution across rounds, and the share-count and option-pool figures an evaluation needs as reference points. - Andy Rachleff and the Wealthfront startup-equity guidance — the widely-cited articulation of why fully-diluted percentage, strike price, and exit scenarios, not headline dollar values, are the terms that determine a grant's worth. - Frederic Kerrest and the early-employee equity literature — the practitioner case that the post-termination exercise window and the AMT exposure on early exercise are the mechanics that most often turn a paper-valuable grant into a real loss. --- - [Next: Dilution](dilution.md) - [Previous: Equity Compensation Types](equity-compensation-types.md)