Last month a founder showed me his "compensation plan." It was a spreadsheet with four salary numbers and the word equity written in a note. That's not a plan — it's a wish. And the bill always arrives later, usually when the best engineer on the team finds out the person hired three months after her got 40% more options for the same title.
Building a compensation plan for startup employees isn't about spreadsheets and market data. It's about fixing the rules of a game before anyone starts playing, so that no one feels cheated six months in. I've helped build these plans at seed stage and at Series B, watched a few collapse, and made a couple of expensive mistakes myself. Here's what I wish someone had told me the first time.
Key takeaways
- Write a pay philosophy on one page before you touch any number — cash-heavy or equity-heavy is a strategic choice, not a default.
- Benchmark by role and stage, then build bands with a floor and a ceiling. Without a ceiling, every negotiation becomes a bidding war.
- Most startups weight fixed salary at roughly 70% of a total package, with variable and equity making up the rest — but the ratio flips by function.
- Vesting (usually 4 years, 1-year cliff) matters more than the grant size for retention. Get it in writing before the offer.
- CEO and founder pay needs the same band logic as everyone else, or the whole plan loses credibility.
How to build a compensation plan for startup employees without losing your team
Start with the philosophy, not the spreadsheet. A pay philosophy is a short document that answers three questions: where do we position against the market (top of market, median, below median plus upside?), how heavy is equity versus cash, and how do we treat people who grow into a new role versus people we hire into it. That's it. One page. Mine fit in four paragraphs the first time I wrote one, and it saved me roughly a dozen awkward conversations over the following year.
Phase 1: define your philosophy before you define a number
A seed-stage company hiring its first ten people cannot compete on cash with a Series C. So don't pretend. The honest version of the philosophy usually reads something like: "We pay at the 50th percentile for base salary and offer equity above what comparable companies offer, because we're asking people to bet on us." That's a trade you can defend in an interview. What you cannot defend is "we pay market" when you don't.
Two decisions matter more than the rest:
- Cash versus equity weighting — do you want people here for years or for the outcome?
- Transparency level — do people know the bands, or do they only know their own number?
On the second, I'll admit I got this wrong. At my first startup we kept bands private, on the theory that transparency would create envy. It did the opposite — people assumed the worst. We eventually published the bands internally and had one difficult week, followed by two quiet years. In my experience, the envy you fear is much smaller than the resentment you create by hiding.
Phase 2: benchmark by role, stage and location
Now the numbers. Pull benchmark data for every role you plan to hire in the next 12 months, and adjust for stage. A backend engineer at a pre-seed company with no revenue and a backend engineer at a Series A with paying customers are not the same job, even if the job description matches. Location still matters too, though the gap between hubs has narrowed a lot compared to a few years ago.
Then build a band. A band has a floor, a midpoint and a ceiling. The floor is what you pay someone still learning the role. The ceiling is the most you'll pay without a promotion. Use the same band for a new hire and an internal promotion, or you'll create a two-tier system without meaning to.
What's a real number to anchor on? Here's a rough order of magnitude I've seen hold across a lot of early-stage teams in Europe and the US, excluding founders:
| Stage | Senior engineer base | Typical equity (4-yr grant) |
|---|---|---|
| Pre-seed (under 5 people) | Below market, often 15–25% under | 0.5%–2% |
| Seed | Near market | 0.25%–1% |
| Series A | At market | 0.1%–0.5% |
| Series B+ | At or above market | 0.05%–0.2% |
Treat those as starting points, not gospel — the spread depends on dilution history and how much of the pool has already been handed out. The takeaway: equity percentages fall fast as you raise, so a 1% grant at seed is a very different animal from 1% at Series B.
How to structure a compensation plan
The most common structure I see is a three-piece package: base salary, a variable component, and equity. Base covers the cost of living. Variable rewards outcomes. Equity rewards the long bet. Where startups break this is by trying to pay everyone the same way — a salesperson needs a heavy variable component, an engineer usually doesn't.
Here's a structure that works for most early teams:
- Base salary banded by role and level
- Variable pay only where the role has a measurable output (sales, growth, sometimes customer success)
- Equity grant sized by level, with a fixed vesting schedule
- A written policy for promotions, refreshers and cost-of-living adjustments
That last item is the one everyone forgets. Without a refresher policy — the practice of granting additional options to people after a couple of years — you'll watch your early employees drift toward the end of their vest and start taking calls from recruiters. Refreshers cost less than replacing a senior engineer, which in my experience runs to several months of lost productivity and a search that stretches longer than anyone planned.
What is a 70/30 compensation plan?
A 70/30 plan means 70% of total target compensation is fixed and 30% is variable, tied to performance against specific targets. It is most common in sales roles, where a rep might earn a base of 70 and a commission potential of 30 on top, but the same logic gets applied to other output-driven positions. The 70/30 split exists to keep the fixed cost manageable while giving strong performers a real upside — which is exactly what a cash-constrained startup needs.
The catch is that 30% variable is only motivating if the targets are achievable and the payout mechanics are clear. I've seen a plan where the variable portion was technically 30% but the quota was set so high that no one hit it for two quarters. The team read it as a 100% fixed plan with a permanent corporate discount, and the top rep left. Set targets you believe at least half the team can hit.
Equity and vesting mechanics
Standard practice is a 4-year vest with a 1-year cliff: nothing vests for the first twelve months, then it vests monthly or quarterly over the remaining three years. Some companies extend to a 5-year schedule or add a 2-year cliff for senior hires — think carefully before you do, because it lengthens the time before an employee has any real skin in the game.
Two things get overlooked. First, exercise windows. If someone leaves, how long do they have to buy their vested options, and can they afford to? Second, the tax treatment — options in the US are typically ISOs or NSOs, and in Europe the structure depends on jurisdiction. Get a lawyer and an accountant to sign off on the plan document before you send a single offer. I once signed off on option grants with a cliff I'd misread, and we spent a week of panicked emails with the lawyer to correct it. Not fun.
What is the typical compensation for a startup CEO?
At seed stage, a first-time CEO usually takes a base salary well below what they could earn elsewhere — often closer to what a senior employee earns than to a market CEO rate — and takes a larger equity stake instead. Typical founder equity at seed (across all founders, before dilution) sits somewhere in the 60–80% range, split among the team. As the company matures, the CEO's cash compensation climbs toward the market rate for the role, and equity dilution does most of the work of "paying" them.
A non-founder CEO hired at Series A or later is a different story: near-market base, a meaningful equity grant (often 2–5% depending on the stage and the candidate's leverage), and a plan that vests over four years. The specific numbers are less important than consistency — if the CEO's compensation doesn't follow the same band logic as everyone else's, the plan loses credibility the moment it becomes visible to the team.
Is 1% equity in a start-up good?
Whether 1% equity is good depends entirely on the stage and the role. At pre-seed, 1% for a senior hire is meaningful but not exceptional. At seed, 1% is a strong offer for a key early employee. At Series B, 1% for anyone outside the executive team would be unusual. By Series C, 1% is typically reserved for top executives.
Spoiler alert: the percentage alone tells you almost nothing. What matters is the fully diluted percentage — calculated after all options, the option pool, and any convertible notes are accounted for — plus the strike price and what the shares actually represent. A 1% grant in a company valued at 30 million is a different bet from 1% in a company valued at 300 million. Ask for the cap table and the fully diluted share count, and get a lawyer to look at the numbers before you accept.
The mistakes that quietly cost you your team
Real talk: the plan is rarely the problem. The communication is. Here's what I've watched go wrong, including once by my own hand.
Under-explaining the equity
I handed an engineer an options grant with a vesting schedule, a strike price and a cliff, and spent four minutes on it. He nodded and signed. Eighteen months later he told me he'd assumed the options were worth roughly their face value at grant, and that he'd built a whole financial plan around it. We fixed it with a proper walkthrough and a conversation about what a liquidity event actually looks like — but the damage to his trust was real.
Every new hire should leave the offer call knowing what options are, what happens if they leave, and what the realistic scenarios are. If you can't explain your own plan in plain English, it isn't ready to sell.
Treating every role the same
A salesperson, an engineer and a designer do not have the same relationship to the equity you offer. Sales can be motivated by cash, engineers often by equity and problem quality, and designers by a mix. Paying everyone the same way is easier to administer and worse for retention. Segment the plan by function, and review it every six months.
Having no policy for the person who grows
The worst quiet cost is the person who joined at a junior level, delivered at a senior level for two years, and never got a formal review. They've stopped asking. They've also started interviewing. Write the promotion and refresher policy down, publish it, and apply it on a fixed calendar — not when someone threatens to resign.
A plan is a promise — write it down
The best compensation plan I ever helped build fit on two pages: a philosophy paragraph, three salary bands, an equity table by level, and a short policy on refreshers and promotions. It wasn't sophisticated. It was consistent, which is the only thing that actually matters over time. People forgive a below-market salary if they understand the reasoning. What they don't forgive is discovering that the rules were never really rules.
So before you write another offer: write the one-page philosophy. Then write the bands. Then write the policy for the people already inside. If the plan doesn't survive contact with the person you're about to hire, it won't survive the next eighteen months either.
