How to Build a Startup Sales Compensation Plan That Drives Growth

Your first sales hire asks about commission. You freeze. You've got maybe fourteen months of runway, no CFO, and a spreadsheet you built at midnight that sort of works. So you do what most founders do: you promise something vague, then spend the next six months quietly hoping the number you said out loud doesn't bankrupt you.

I've been on both sides of this. I've designed comp plans as a founder and I've watched plenty of them fall apart. Here's the uncomfortable truth about a startup sales compensation plan: it is not an HR document. It's a cash-flow forecast wearing a costume. Every dollar of variable pay you promise is a dollar you've committed before you've seen the revenue it's supposed to generate.

So let's build one that survives contact with reality.

What a startup sales compensation plan actually has to do

Most founders treat the plan as a recruitment tool. It is, partly. But its real job is narrower: it should make your sales economics legible. If you can't look at your plan and say "here's what a rep costs me per dollar of ARR they close," you don't have a plan. You have a promise.

Three things a working plan must accomplish:

• Fund the hire without starving the runway. Variable pay should be self-funding wherever possible.
• Point the rep at the right behavior: new logos versus expansion, annual contracts versus monthly, small fast deals versus long enterprise ones.
• Survive a bad quarter. If a slow Q1 makes your best rep quit because they can't pay rent, the plan failed.

The question to answer before you touch a spreadsheet

How predictable is your sales cycle? Be honest. If you're closing deals on founder relationships and referrals, your cycle isn't predictable—it's luck with a pattern. In that case, a heavily commission-weighted plan is a trap. You'll be paying variable comp for revenue that would have arrived anyway, and you'll have no idea whether the rep caused it.

I made this mistake. My first hire got a 50/50 split because I'd read that aggressive variable pay "aligns incentives." Six months in, she'd closed four deals. All four came from my network. I'd paid roughly $18,000 in commission on revenue I'd sourced myself, and I learned nothing about whether she could sell. That's not alignment. That's an expensive education.

How to create a sales compensation plan, step by step

Here's the sequence I use now, in order. Skip a step and you'll rebuild the whole thing in four months.

Step 1: model the total cost before you name a number

Start from what you can afford, not from what the market pays. Take your annual sales budget, divide by the number of reps you want, and you get your ceiling per head—fully loaded. That includes base, variable at target, payroll taxes, any tooling, and a fudge factor for the quarter where nobody hits quota.

A realistic fudge factor: assume 60–70% of reps hit target in any given quarter. That is not pessimism. That is how quota distributions actually behave. If your model assumes everyone hits 100%, your budget is fiction.

Step 2: set OTE and pick your split

OTE is the total a rep earns if they hit quota exactly. The split is how that OTE divides between guaranteed base and at-risk variable. Say your OTE is $120,000 with a 70/30 split: $84,000 base, $36,000 variable.

Which split? Use this rough guide:

Split	Base / Variable on $120k OTE	Best for
90/10	$108k / $12k	Pre-seed. Founder-led sales, unpredictable cycle, rep is still learning the pitch.
80/20	$96k / $24k	Seed stage. Repeatable motion emerging, but you're still figuring out what works.
70/30	$84k / $36k	Series A. Playbook exists, quota is defensible, you want real upside pressure.
60/40 or 50/50	$72k / $48k	Mature, high-volume, short-cycle sales. Rarely right for an early startup.

Notice the pattern: the more predictable your revenue engine, the more you can shift into variable. Early on, you're buying certainty for the rep because you can't offer certainty about the pipeline.

Step 3: set the quota from the OTE, not the other way around

Here's the calculation nobody explains well. You want the rep's quota to generate enough gross margin to cover their fully loaded cost plus a return. A common approach: quota should be roughly 4–6x the rep's OTE in new ARR for a SaaS business with healthy margins.

So a rep on $120,000 OTE carries a quota somewhere between $480,000 and $720,000. That range is wide on purpose—it depends on your gross margin and how much of the sale is yours versus theirs.

The catch? At seed stage you often can't defend any quota. Which is why you should consider paying a floor: a guaranteed minimum variable payout for the first two or three months while the rep builds pipeline. It costs you a few thousand dollars and buys you a rep who doesn't panic in week six.

Step 4: design accelerators and caps

Accelerators are where reps make real money. A typical structure: 100% of commission rate up to quota, then 1.5x or 2x the rate on everything above.

Should you cap earnings? My opinion, and I'll defend it: no cap, but do cap the plan's exposure by capping quota attainment crediting at, say, 200%. An uncapped plan with a runaway rep is a good problem to have—you're paying out on revenue you actually booked. A cap that punishes your best performer in a record quarter is how you lose them to a competitor in the following month.

What does a 70/30 split in sales compensation mean?

A 70/30 split means 70% of a rep's on-target earnings is fixed base salary and 30% is variable, paid on performance against quota. On a $100,000 OTE, that's $70,000 guaranteed and $30,000 at risk.

What matters is what the split signals. A 70/30 structure tells the rep: "we believe the pipeline is real enough that you can earn this variable reliably, and we want you pushing hard for it." It also tells them your revenue motion is mature enough that quota is a fair measure of effort.

The practical consequence founders miss: a 70/30 split raises your fixed cost base by roughly 17% compared to an 80/20 on the same OTE. That's real money off your runway. Only take on that fixed commitment when you can name three deals in the pipeline that a rep—not you—could plausibly close.

What is an 80/20 sales compensation plan?

An 80/20 plan puts 80% of OTE in base and 20% in variable. On $100,000 OTE: $80,000 base, $20,000 commission at target.

This is the plan I'd default to for most seed-stage startups, and I'll say why plainly. At that stage you are not hiring a closer. You are hiring someone to run experiments on your behalf: which segment responds, which pitch lands, which objection kills deals. An 80/20 gives them the psychological safety to run those experiments without a mortgage payment hanging on the outcome of every call.

It also protects you. Your variable cost is lower, which means a bad quarter hits your bank account less hard. And when you eventually raise and move to 70/30, the transition feels like a promotion rather than a pay cut.

What about 90/10, or no variable at all?

At pre-seed, a 90/10 or even a base-only plan is defensible. You're hiring a generalist who will do outbound, run demos, write the follow-up emails, and probably fix the CRM. Paying them mostly on commission for a motion you haven't defined yet is unfair to them and uninformative for you.

Here's the thing: several founders I know went base-heavy early, then converted to a proper variable structure once they had a repeatable playbook. Their reps stayed. The founders who went 50/50 on hire number one mostly had to re-hire within a year.

What is the typical compensation for a startup CEO?

Early-stage CEO pay varies enormously, but the pattern is consistent: founder-CEOs at pre-seed and seed typically take a modest salary, often well below market for their experience level, and hold the bulk of their compensation in equity. At Series A and beyond, salaries climb toward something resembling market rate for an operator of that scope, with equity still forming a meaningful part of the package.

Why does this belong in an article about sales comp? Because your own salary sets the ceiling for what you can credibly pay anyone else. If you're taking a reduced salary, a rep on a $120,000 OTE is a bigger relative commitment than the number suggests. Founders who forget this end up with a sales hire earning more than them and a weird, unspoken resentment that poisons the relationship.

The other reason: if you plan to hire a VP of Sales, their package has to be benchmarked against a market you can't see from the inside. Talk to two or three founders who've made that hire recently. Not for exact numbers—for the shape of the deal, the split, and what went wrong.

The mistakes I made, so you can skip them

Three failures worth naming.

I announced a plan verbally. No document. Six months later my rep and I remembered the accelerator differently. We were both acting in good faith and we were both wrong. Write the plan down, date it, have both parties sign it. One page is enough.

I paid commission on bookings, not collections. One customer signed an annual contract, then churned after two months. The commission was already out the door. Now I pay on collected revenue, or at minimum split the commission across the contract term.

I never modelled the plan's total cost as a scenario. Here's what I do now, and it takes twenty minutes: build a small table with three columns—"rep hits 50% of quota," "rep hits 100%," "rep hits 150%"—and put your fully loaded cost next to the gross margin each scenario generates. If the 50% column makes you wince, your base is too high for your runway. Run the exercise before you sign, not after.

How often should you revise the plan?

Twice a year, and only at the boundary of a period. Never mid-quarter, and never retrospectively for deals already closed. Retroactive changes are the fastest way to teach a rep that your numbers aren't real—and once they believe that, they stop trusting every target you set.

If the plan is genuinely broken mid-period, fix it going forward and eat the cost of the current quarter. It stings. It's cheaper than the replacement hire.

The plan you write in month one is a guess. The plan you write in month twelve, after watching a real human try to sell your product, is a decision. Give yourself permission to be wrong early—just make sure you've written the guess down, so you can see exactly where it broke.