A startup offer and an established one are different propositions rather than two versions of the same thing. Value the equity honestly by asking what percentage of the company it represents, at what valuation, over what vesting schedule, and with what exercise terms — then decide whether you would take the role at zero equity value. If the answer is no, the equity is doing work it cannot be relied on to do.
Two different propositions, not two versions of one
The instinct is to put both offers on a spreadsheet, add up the numbers, and pick the larger total. That treats them as the same kind of thing measured in the same units, and they are not. One offer is mostly cash you will certainly receive; the other is less cash plus a claim on an outcome that may or may not happen.
Adding an expected equity value to a salary and comparing the sum to another salary produces a number with two very different kinds of certainty inside it. The arithmetic works and the comparison does not, because a dollar you will definitely receive and a dollar that arrives with a thirty percent probability are not interchangeable however carefully you weight them. Certainty is not a discount you can apply.
So the useful approach is to hold the two apart. Compare the certain parts against each other, understand what the uncertain part actually is, and then make a judgment about how much uncertainty you want. That is a decision about your own circumstances rather than an arithmetic problem with a right answer.
How to value the equity honestly
Start with what you have actually been offered, which is usually a number of shares or options rather than a percentage. A share count on its own is meaningless without the total outstanding, and employers sometimes quote the count precisely because it sounds substantial on its own. Fifty thousand shares means nothing without a denominator.
Convert it to a percentage of the company on a fully diluted basis, and be specific about fully diluted — the denominator should include all outstanding shares, all options granted and reserved, and any convertible instruments. A percentage calculated against a smaller denominator overstates your position, sometimes considerably. Ask specifically for the fully diluted figure rather than accepting the first one offered.
Then apply the current valuation to get a face figure, and treat that figure with appropriate skepticism. A valuation set at the last funding round is a price somebody paid for preferred shares with protections attached, and your common shares do not carry those protections. In an outcome below the last round’s price, preferred holders are frequently paid first and common holders receive substantially less than the headline arithmetic implies.
The questions that actually price it
Four questions get you most of the way, and they are ordinary things to ask during a hiring process. Nobody serious is offended by them.
What percentage of the fully diluted company does this grant represent? What was the valuation at the most recent round, and when was that round? What is the vesting schedule, including any cliff? And what is the preference stack — how much has to be returned to investors before common shareholders see anything?
That last question is the one candidates rarely ask and the one that most affects the answer. A company with a large accumulated liquidation preference can be sold for a substantial sum with common shareholders receiving very little, and the arithmetic on that is not something you can work out from the outside without being told. That is why it has to be asked.
An employer who answers all four readily is being straight with you. An employer who cannot or will not answer is telling you something about how information will flow once you are inside. Treat the refusal as part of the offer.
The exercise problem
Options are not shares and the difference has caught a very large number of people. An option is a right to buy at a fixed price, and exercising it costs money — sometimes a great deal of money, and frequently a tax bill on top in the year you exercise, on a gain you cannot sell to fund it. People are caught by that combination every year.
The deadline is the part that cannot be recovered from. In most standard plans, leaving the company starts a window — commonly around ninety days — in which you must exercise or forfeit. People discover this at the worst moment, decide they cannot fund the exercise, and lose vested equity they had already earned.
Some companies have extended that window to several years, and where they have, it is a genuinely meaningful benefit worth asking about explicitly. It is also a good proxy question: a company that has thought about post-termination exercise windows has thought about its employees’ actual position rather than only its own. It is a useful proxy for a great deal else.
What the startup genuinely offers
It is worth stating the real case rather than only the risks. Scope arrives faster at a small company, because there are fewer people and the work has to be done by somebody. Two years at a growing startup can produce experience that takes six years to accumulate in a structured organization.
Proximity to decisions is real too. You see how the whole business works rather than one function of it, and that understanding is portable in a way that specific technical experience sometimes is not. It travels with you to any employer.
And the upside, while unlikely, is genuinely asymmetric. Most outcomes are modest and a small number are life-changing, which is a distribution worth being clear-eyed about rather than either dismissing it or planning around it. Both errors are common and the second is worse.
What the established employer genuinely offers
More of the total is cash you will actually receive, which compounds in a way an uncertain claim does not. A higher base raises every future percentage increase and every future offer benchmarked against it, and that effect runs for the rest of your career rather than until an exit that may never come. Cash compounds on a schedule nobody has to approve.
Benefits are usually better and are worth more than people count. Retirement matching, health coverage, paid leave and disability insurance are real compensation, and the gap between a mature benefits package and a thin one can run to five figures a year. Most people never price it.
There is also structure — defined bands, published levels, a promotion process, and a name that is legible to every future employer. None of that is exciting and all of it reduces variance, which is the thing you are actually choosing between. Variance, not value.
The risk that is not about the company
The risk people underweight is not that the company fails. It is that it neither fails nor succeeds, and continues for six years in a state where your equity is worth roughly what it was on your first day while your cash compensation stayed below market that whole time. Nothing dramatic happens in that story, which is why nobody plans for it.
That outcome is far more common than either the failure or the exit, and it is the one nobody plans for. The cost is not the equity, which was always uncertain; it is the accumulated difference in cash compensation across those years, which is certain and compounds every year. The equity was always uncertain; that gap never was.
Run that scenario before deciding. If the salary gap is $20,000 a year and the realistic horizon is five years, the certain cost is $100,000 before compounding, and the equity has to clear that bar in expectation before it is even a fair trade. Run that number rather than estimating it.
How to decide
The test that cuts through all of it is a single question: would you take this job if the equity turned out to be worth nothing? Consider the role, the people, the learning, the cash compensation and the trajectory, and answer honestly rather than aspirationally. The honest answer usually arrives immediately.
If the answer is yes, take it and treat the equity as a genuine possibility rather than a plan. If the answer is no, the equity is being asked to carry the decision, and equity is the one component that cannot be relied on to carry anything at all. That is not pessimism; it is what an option is.
That test also protects you from the version of this decision that goes worst: accepting a role you were lukewarm about because a spreadsheet made an uncertain number look large. The spreadsheet was never the problem — treating one column of it as certain was.
Common questions
Can I just compare the totals?
No. One offer is mostly cash you will certainly receive; the other is less cash plus a claim on an uncertain outcome. Adding an expected equity value to a salary mixes two kinds of certainty in one number.
How do I value the equity?
Convert the share count to a percentage of the fully diluted company, apply the most recent valuation, and treat the result skeptically — that round priced preferred shares with protections your common shares do not have.
What four questions should I ask?
What percentage of the fully diluted company; the valuation and date of the last round; the vesting schedule including any cliff; and the preference stack — what must be returned to investors first.
Which question matters most?
The preference stack. A company can be sold for a substantial sum with common shareholders receiving very little, and you cannot work that out from the outside.
What is the exercise problem?
Options cost money to exercise, often with a tax bill on a gain you cannot sell to fund — and leaving usually starts a window, commonly around ninety days, after which vested equity is forfeited.
What does a startup genuinely offer?
Faster scope, proximity to how the whole business works, and a genuinely asymmetric upside that is unlikely rather than impossible.
What is the underweighted risk?
Not failure. It is the company neither failing nor succeeding for six years while your cash compensation stayed below market — a certain, compounding cost against an uncertain gain.
What is the deciding test?
Would you take the job if the equity turned out to be worth nothing? If no, the equity is being asked to carry the decision, and it is the one component that cannot carry anything.