How startup equity actually makes money (or does not)

Startup equity makes money in exactly one way: the company gets acquired or goes public, and your shares are worth more than you paid to exercise them. Until that exit, your equity is a paper claim, and three forces decide what the claim is actually worth: how much you get diluted across future rounds, how much the investors get paid before you through liquidation preferences, and whether there is an exit at all. Most grants resolve to zero, a handful pay a salary's worth, and a small number change your life. This post walks the real math, with numbers, and marks every worked example as illustrative.

None of this is a reason to avoid equity. It is a reason to understand it before you weight an offer on it. The roles.cc board sorts roles by how recently the company raised, because the round you join at sets the price of your shares and the dilution ahead of you. Both feed straight into the math below.

What does it actually take for equity to pay out?

Three things have to line up, in order. Miss any one and the equity is worth nothing.

1. 01There has to be an exit. An acquisition or IPO that turns shares into cash or tradable stock. No exit, no payout, no matter how good the product was.
2. 02The exit price has to clear the preference stack. Investors get their money back first. If the company sells for less than what was raised, common shareholders (you) can get zero even on a sale that looks like a win.
3. 03You have to own the shares. That means you stayed long enough to vest, and you exercised your options (paid the strike price) at some point, often with real money and a real tax bill.

Notice that strong product and happy customers are not on this list. Equity is a financial instrument, not a performance review. A great company that never exits pays the same as a failed one: nothing. We cover the join-side of this decision in is a startup job worth the risk.

How much does dilution shrink your slice?

Every time the company raises a new round, it issues new shares. Your share count stays the same, but the total grows, so your percentage falls. This is dilution, and it is normal. A typical priced round sells 15 to 25 percent of the company, plus an option pool top-up that dilutes everyone again.

Say you join after a seed round with a grant worth 0.5 percent of the company (illustrative, not advice). Walk it forward through a Series A, B, and C, each selling roughly 20 percent, and your slice erodes like this:

So a 0.5 percent grant is really closer to 0.26 percent by the time a late round closes (illustrative, not advice). Dilution is not theft. The new money usually grows the pie faster than it shrinks your slice, so 0.26 percent of a $500,000,000 company beats 0.5 percent of a $40,000,000 one. The point is to do the math on the diluted number, not the headline one. For how grant sizes differ by stage, see how much equity a startup engineer gets by stage.

What is a liquidation preference, and why can it zero you out?

This is the part most engineers never see until it bites. When investors put in money, they almost always get a liquidation preference: the right to be paid back first if the company sells. A standard term is 1x non-participating, meaning each investor gets back either their money or their ownership percentage of the sale, whichever is larger. Common shareholders split whatever is left.

Here is why it matters. Imagine a company that raised $80,000,000 total and sells for $100,000,000. That looks like a good outcome. But investors take their $80,000,000 off the top first. Only $20,000,000 is left for common shares (illustrative, not advice). If you owned 0.26 percent of the company, your share of that residual is about $52,000, not the $260,000 the headline price implied.

The sale price is not what you get paid on. The price minus the preference stack is.

Now make the sale price $70,000,000 instead. Investors are owed $80,000,000. They take the whole $70,000,000, and common shareholders get zero. The company sold for $70,000,000 and your equity was still worth nothing. This is how a sale that reads like a success in the press pays employees nothing. Participating preferences and stacked multiples (2x, 3x) make this worse, which is why the terms on the cap table matter as much as the valuation. We list these in startup equity: seed, A, B, what to ask.

What is a single grant actually worth, on average?

The honest answer uses expected value: the payout in each outcome, weighted by how likely that outcome is. Venture outcomes are not a bell curve. They are a power law. Most companies return little or nothing, and a few return almost everything. Any average is dragged up by rare big wins you probably will not land.

Take that diluted 0.26 percent grant and put rough probabilities on the exits (illustrative, not advice):

Weight those out and the expected value is roughly 0.20 times $1,040,000 plus 0.05 times $5,200,000, or about $468,000 in expectation (illustrative, not advice). That is a real number. It is also a number with a 75 percent chance of being zero. The expected value is genuine, but you cannot spend an expectation. You get one draw, and the most likely single result is nothing. This is the gap between equity as a portfolio bet (sound) and equity as your personal retirement plan (fragile).

What about the strike price and taxes?

The payouts above are gross. To actually own the shares, options holders pay the strike price to exercise, and the IRS often taxes the gap between strike and current value at exercise, before any sale. On ISOs, that gap can trigger the alternative minimum tax, a cash bill on gains you cannot yet sell. People have paid five-figure tax bills on paper gains that later went to zero.

This is why the mechanics matter as much as the grant size. Whether you have ISOs, NSOs, or RSUs, what the vesting cliff is, and how long you have to exercise after you leave all change the real number. We break those down in how stock options and vesting work. If your offer is mostly equity, read it before you sign.

So how should you weight equity in an offer?

Treat equity as a lottery ticket with good odds attached to a job you would take anyway, not as deferred salary. Concretely:

• Live on the cash. Make sure the base salary alone is a number you can accept. See senior software engineer salary in SF and NYC for 2026.
• Discount the equity hard. A reasonable mental haircut is to value a grant at a small fraction of its headline number, because of dilution, preferences, and exit odds.
• Ask the terms, not just the percent. Total raised, preference multiple, participating or not, and your own diluted ownership. These decide whether a sale pays you.
• Join earlier, not just bigger. A fresh round means a lower share price and more dilution still ahead, but also more upside if it works. Funding recency is the cleanest read on where a company is in this cycle, which is why we sort the board by it.

Equity is worth having. It is not worth overpaying for in foregone salary, and it is not worth signing without understanding the preference stack sitting above you. To pressure-test a full offer, walk through how to evaluate a startup job offer.