A recruiter says the offer is “$340,000.” Sit with that number for a second, because it’s doing a lot of quiet work. It adds your base salary, a bonus you haven’t earned yet, and one-quarter of a four-year stock grant valued at today’s share price, then presents the sum as if it lands in your account every year. The year you actually live through usually pays something quite different.
Pay at tech companies comes in four pieces, and they behave nothing alike. Knowing how each one pays out over time is what separates the offer that looks bigger from the one that pays more.
The four parts, and how each one really pays
Base salary is the only piece you fully control once you sign. It hits your account every two weeks, it doesn’t depend on the stock market or a manager’s rating, and it’s the number the rest of your financial life keys off. Treat it as your floor.
The bonus is a target, usually quoted as a percentage of base: 10% at many companies, 15 to 20% at some, higher for senior and staff roles. Target means midpoint, not promise. It’s tied to company performance and your rating, it’s often prorated in your first year because you didn’t work the whole cycle, and in a bad year it shrinks. Plan around the target, but don’t spend it before it clears.
Equity is where offers stop being comparable. At a public company it’s RSUs, restricted stock units, granted either as a share count or a dollar value the company converts to shares at signing. The figure the recruiter quotes assumes the stock price never moves for four years. It will move. And the schedule that controls when those shares become yours varies enough between companies to flip which offer is better.
The sign-on bonus is a one-time payment, sometimes paid in full up front, more often split across your first two years. It frequently carries a clawback: leave inside twelve or twenty-four months and you repay a prorated chunk. Sign-on money exists to paper over a gap somewhere else in the package, which is a clue worth reading.
Why “annual equity” is a number nobody actually receives
Take a grant quoted at $200,000 over four years. The recruiter calls that $50,000 a year. Whether you ever see $50,000 in a given year depends entirely on the vesting schedule.
The common public-company schedule is four years with a one-year cliff: nothing vests until your first anniversary, then 25% lands at once, and the rest comes monthly or quarterly after that. Quit at month eleven and you walk away with zero equity. Make it past the cliff and the income is reasonably smooth.
Amazon runs a different schedule, and it’s the textbook case for why the average misleads. Its RSUs vest 5% after year one, 15% after year two, then 20% every six months through years three and four, so 40% and 40%. To keep your first two years from looking thin against competitors, Amazon pays cash sign-on bonuses in years one and two that taper off exactly as the stock ramps up. The “average annual” figure is only real if you stay the full four years. Leave after two and you collected the two small slices plus the sign-on, well under the quoted number.
| Year | Standard 25/25/25/25 | Amazon-style 5/15/40/40 |
|---|---|---|
| 1 | $50,000 | $10,000 |
| 2 | $50,000 | $30,000 |
| 3 | $50,000 | $80,000 |
| 4 | $50,000 | $80,000 |
Same $200,000 grant, wildly different cash flow. If you’re the kind of engineer who changes jobs every two to three years, the back-loaded version pays you a fraction of what the front-loaded one does, and the single total-comp number hides that completely.
Startup equity is a separate animal. It’s usually stock options with a strike price, not RSUs, and it vests over four years with monthly increments after a one-year cliff. The dollar value attached to it comes from the company’s most recent 409A valuation or, worse, a number the founder hopes the next round will hit. You can’t sell it. You may owe tax (AMT on incentive stock options) when you exercise, years before there’s any liquidity. Counting startup options at face value next to public RSUs is the most expensive mistake people make reading offers.
The year you’ll actually live in beats the four-year average
Most engineers don’t stay four years. Median tenure at large tech companies sits closer to two or three, which means the steady-state average is a destination you’ll probably never reach. Two numbers tell you more than the headline.
Year one is almost always distorted. A sign-on bonus pushes it up; an equity cliff or a back-loaded schedule drags it down. Whatever the recruiter’s single figure says, your first twelve months will look nothing like it.
Steady state, roughly years two and three, is the truer read, but only if refresher grants keep your equity topped up. Refreshers are additional stock grants companies hand out annually or at promotion, and they’re the reason a package doesn’t collapse in year four when the original grant finishes vesting. That drop has a name in offer-letter folklore, the vesting cliff, and it catches people who never asked about refresh policy. Some companies refresh generously and predictably. Others stay vague on purpose. Ask before you sign, because the answer changes your three-year math.
How to compare two offers without fooling yourself
Stop comparing headline to headline. Build a year-by-year cash flow for each offer, for the number of years you realistically expect to stay, and add up base plus the bonus you’d actually expect plus the equity that actually vests plus any sign-on. Our total comp calculator lays this out for you, but a spreadsheet works fine. The point is to see the years side by side instead of trusting one blended figure.
When you value the RSUs, use the current share price, then stress-test it. Ask what the offer looks like if the stock falls 30%, because public stock you can sell the day it vests, while a private option is locked up indefinitely. Discount private-company equity hard. It might be worth a fortune; it might be worth the paper it’s printed on. Weighting it the same as liquid stock is how people talk themselves into the wrong job.
A few things are worth pinning down with the recruiter before you decide:
- The exact vesting schedule, month by month, not “four years.”
- Whether the sign-on is one payment or split, and the clawback terms.
- The refresh policy: typical refresher size and when the first one lands.
- Whether the equity figure is a share count or a dollar value, and the price it was struck at.
Those four answers reshape the year-by-year table more than another $10,000 of base will.
Where the real money leaks
The costly errors cluster in a few places. Treating a startup’s hopeful valuation as cash. Forgetting the quoted equity assumes a flat stock price, when a grant struck at a peak two years ago can be worth half today. Ignoring the year-four cliff and the refresh question. Banking the target bonus as if it were guaranteed base.
The lever you push in a negotiation should match the year you care about. If you expect to leave in two years, fight for base and sign-on, since back-loaded equity won’t reach you. If you’re planting roots and believe in the stock, push for a larger grant and a clear refresh commitment. Build the year-by-year table first, then negotiate the offer that pays well in the years you’ll actually be there.
Line up the rest of your job search:
