Four offers out. Three signed. Then the fourth candidate asks the one question that can blow up a seed-stage team: "What package did you give the first engineer?" And you realize you never wrote down a rule for that question, because until now nobody asked it.
That moment is exactly why you need a startup compensation plan for early employees built before you make the first offer, not after. I've watched founders wing it and I've watched founders build bands on a spreadsheet at 11pm. The second group sleeps better. Not because their numbers were perfect — they weren't — but because they could explain them.
Here's what actually works, what doesn't, and where the standard advice quietly falls apart.
Key takeaways
- Build the cash/equity grid before your first hire, not after. Retrofitting fairness is expensive and visible.
- Early employees typically take a cash discount in exchange for meaningful equity — but the discount has to be negotiated, not assumed.
- Document bands by role and level. A verbal promise is a future lawsuit in slow motion.
- Equity percentages shrink fast as you hire. A "big" number for employee #4 can look small by employee #15 — say so out loud during the offer.
- The cofounder split and the early employee package are two different problems. Do not solve them with the same spreadsheet.
How to build a compensation plan for early employees that survives the fourth hire
Compensation at seed stage is a negotiation between two fears. The company's fear: running out of runway because salaries ate the round. The employee's fear: taking a below-market salary for equity that turns out to be worth nothing.
Every plan is a truce between those fears. The mistake almost everyone makes is trying to settle that truce per candidate, in the moment, on a call. That works for hire #1. It becomes chaos by hire #5.
The three columns you actually need
Before any offer goes out, you need three columns per role, per level:
- Market cash — what this person would earn at a mid-size company doing similar work. You can estimate this from conversations, your own network, public job postings. Don't guess wildly; get three data points.
- Offered cash — your actual number. Typically a discount, sometimes 10–20%, occasionally more for very senior early hires where equity is the whole pitch.
- Equity grant — expressed as a percentage of fully diluted shares, with a vesting schedule and cliff written next to it.
Three columns. One row per role. That's the whole system at its core. Everything else is governance.
Why the first hire is a trap and the fourth one is the real test
Hire #1 is easy. You're excited, you overpay slightly, you give a generous number. Nobody's watching.
Hire #4 is where it breaks. By then employees #1 through #3 know each other. They talk. If #3 got a materially better package than #2 for similar work and similar start dates, you will hear about it — usually not from the person who got less, but from the person who got more and feels awkward about it.
The fix is not secrecy. The fix is a written band that you can point to. "Here's the band for your level. You're at the top of it because of X." That sentence does more for retention than a 5% bump.
What is the 80/20 rule for startups?
The 80/20 rule for startups is a rough planning heuristic: allocate about 80% of your compensation budget to cash and about 20% to equity — or, in a different framing, expect roughly 80% of the value of any given offer to come from salary and 20% from equity, weighted by how much risk the employee is taking. It's a sanity check, not a law. At pre-seed with no revenue, the split tilts hard toward equity because you cannot afford otherwise. At Series A with real traction, it tilts back toward cash because candidates can get cash elsewhere.
I've seen founders treat 80/20 as gospel and hand out 0.4% to a senior engineer who then left in seven months. I've seen the opposite: overpaying in equity at the seed stage, then having nothing left for the Series A hires who actually need it. The rule is a starting frame. The real question is: what does this specific person need to say yes and stay?
Should cofounders be 50/50 or 51/49?
For two cofounders, 51/49 is generally the safer default, because it creates a clear tiebreaker on decisions that cannot wait. A 50/50 split works only if you have a written deadlock-resolution mechanism — a third-party advisor with a casting vote, a defined escalation path, or a buy-sell clause that triggers on sustained disagreement. Without one of those, 50/50 is a coin flip waiting to happen at the worst possible moment.
But here's the thing nobody tells you: the percentage matters less than the vesting. A 60/40 split with a four-year vest, a one-year cliff, and a repurchase right on early departure is far healthier than a 50/50 with no vesting at all. I've watched a cofounder walk out at month three with 25% of the cap table because nobody wrote a vesting schedule. That company died in the next round.
Split by contribution and commitment, document it, vest everything. The exact number is secondary.
How much equity should early employees get?
Here's the honest range for a seed-stage company, expressed as a percentage of fully diluted shares:
| Employee number | Typical equity range | Notes |
|---|---|---|
| First engineer / first key hire | 0.5% – 2% | Larger end if you're pre-product and they're taking real cash risk |
| Hires 2–5 | 0.25% – 1% | Declines roughly 10–20% per hire in the same role |
| Hires 6–10 | 0.1% – 0.5% | Specialist senior roles can still command the top end |
| Hires 11–20 | 0.05% – 0.25% | Very role-dependent; leadership hires may exceed this |
These are ranges, not targets. The upper bound of each band should go to someone doing something you cannot easily replace. The lower bound to someone doing work that's genuinely important but has a clear market of people who can do it.
One structural note: the option pool should be sized to cover the next 18–24 months of hiring, not the next five years. Over-sizing the pool dilutes founders unnecessarily. Under-sizing it forces an ugly pool expansion mid-round that dilutes everyone a second time.
Is 1% equity in a start-up good?
Yes, 1% is a solid early-employee package — for the first handful of hires at a pre-seed or seed-stage company taking below-market cash. At later stages, 1% for a non-executive hire is generous and unusual. At employee #15 at Series A, 1% would be exceptionally high.
"Good" depends entirely on context:
- At what stage? 1% at seed is normal. 1% at Series C is not.
- What's the vesting? Four years with a one-year cliff is standard. Anything shorter is a red flag about the company's governance.
- What's the strike price? A low strike price today is worth much more than a low percentage tomorrow.
- What's the cash trade-off? If you're taking a 40% salary cut for 1%, that math only works if the outcome is very good.
I'd rather see a candidate negotiate vesting terms than fight for another 0.1%. Vesting terms are enforceable today. The percentage is a bet on the future.
The legal checklist that nobody puts in the offer letter
Before you send any offer, confirm:
- Vesting schedule is in writing, with cliff, monthly or quarterly cadence, and acceleration terms if any.
- Exercise window post-departure is defined — 90 days is the standard default and it is aggressively short for employees who can't afford to exercise. Many companies now extend to 5 or 10 years; it costs you nothing meaningful and it removes an enormous retention risk.
- 409A valuation is current. If it lapsed, your strike price is a legal problem.
- Pool size is authorized by the board, not just implied.
- The offer letter doesn't promise anything the plan document contradicts. This is where the $11,000 legal fees come from — one line in a letter that the plan document overrides.
Get a lawyer to read the offer letter template once. It's a few hundred dollars and it prevents a few thousand dollars of cleanup later.
What to do when your early hires start comparing numbers
Assume they will. Employees talk, and you should want them to — a company where nobody discusses comp is a company where nobody trusts the comp.
The move is not to prevent comparison. It's to have an answer ready when it happens:
- Have a written band per role and level, shared with employees who ask.
- When someone is below band, plan a correction. Not necessarily this quarter, but on a schedule.
- Never lie about what you offered someone else. If you can't tell the truth, say "I'm not going to discuss another person's package" — which is different from lying.
Fairness isn't a feeling. It's a documented structure that survives the moment someone discovers a number they didn't expect.
The part people skip
The best compensation plan I've seen at a seed-stage company was two pages long. One page of bands by role. One page of vesting and governance rules. It took a founder and a fractional HR lead an afternoon to write. It's the reason that company made forty offers without a single negotiation blowing up a relationship.
The plan doesn't need to be perfect. It needs to exist before the conversation, so the conversation becomes about the person rather than about your panic.
And the fourth candidate asking about the first engineer's package? That's not a problem. That's a sign you have a system worth explaining.
