Why Your First AE Comp Plan Should Not Have a Variable Component (And What to Use Instead)

Most founders building their first AE comp plan spend their energy on the base/variable split. Sixty-forty, fifty-fifty, seventy-thirty, they debate the ratio like it’s the decision that matters. It isn’t. The decision that matters happened earlier, when they wrote a quota into the plan without having any real basis for the number.

The Quota Is the Problem, Not the Ratio

Variable pay only works if it’s measuring performance against something real. A quota is a claim: this rep should be able to close $X in a given period, because that’s what the role and the market support. At $3M-$15M ARR, before a dedicated AE has ever run a full cycle, founders don’t have that claim. They have a guess dressed up as a target.

Consider what a founder actually knows at this stage. They know roughly how long deals have taken when they closed them personally, but not whether that’s typical or survivorship bias from the deals that worked. They don’t have a stable win rate, because the sample size is small and the founder’s own selling behavior (which prospects they chase, which they let go cold) has been shaping that number invisibly. They don’t have segment-level ACV data, because pricing and packaging have probably shifted twice in the last two quarters. A quota built on top of that isn’t aggressive or conservative. It’s arbitrary. You’re compensating someone against a number that doesn’t yet exist as a measurable fact about your business.

This is true even at companies that think they’re doing it the standard way. Scale Venture Partners frames quota-to-OTE ratio as a sales-efficiency lever, not a target to hit for its own sake. Their own published data puts Net Sales Efficiency at 0.59 for a 4x quota-to-OTE ratio, 0.75 at 5x (their stated “typical company” baseline), and 0.91 at 6x. Most early-stage companies can’t actually hit that 5x starting point with a confident quota, because they haven’t accumulated the cycle-length and win-rate data needed to size one. That’s not a knock on the benchmark. It’s evidence the underlying quota logic is softer at Series A than the benchmark assumes once a company matures into it. Everstage’s 2024 benchmarking puts median OTE for a SaaS AE at roughly $190,000, up from $167,000 in 2022, with something close to a 53:47 base-to-variable split, citing data from The Bridge Group. That structure gets copied constantly because it’s the template everyone has seen, not because anyone checked whether the underlying quota logic holds at Series A.

Why “Just Pay More Base” Doesn’t Fix It

The instinct, once a founder notices the quota problem, is to lean the plan toward base pay and call it conservative. That doesn’t solve anything, it just relabels the guess. Falcon Incentives’ startup comp guide recommends a 50/50 or 55/45 base-to-variable split for an Account Executive, reserving heavier base weighting for the SDR role, where the job is pipeline generation rather than closing. Even at that AE split, the plan still expects a meaningful variable component that rewards outperformance. Vouris reaches a similar diagnosis from the opposite direction: they recommend keeping AE base pay at 60-70% of OTE specifically because early-stage data is too thin to project performance reliably, moving toward heavier variable comp only once a company has matured and has clean sales data. Different numbers, same underlying point: don’t over-index on variable pay before you have the data to size it fairly. That expectation runs into a wall the moment there’s no valid quota to measure outperformance against. You can’t reward outperformance relative to a target you invented. Shifting more of the plan toward base doesn’t remove the invented target, it just makes it a smaller invented number attached to a bigger guaranteed check. The plan is still measuring the rep against fiction, just at a lower dollar amount.

Treat the First AE’s Job as Data Collection

The reframe that actually works: the first AE’s real job isn’t to hit a number, it’s to build the data your business doesn’t have yet. Every deal they run should be turning implicit founder knowledge (how a cycle actually unfolds, why prospects disqualify, which objections recur) into documented, repeatable process. That’s the asset. The quota comes later, once that asset exists.

Practically, this means paying against process artifacts instead of closed-won ARR: documented cycle stages with real dates attached, win/loss reasons logged for every closed deal (not just the wins), and pipeline coverage built to a defined multiple. These are milestone or discretionary bonuses, not commission, because commission implies a rate against a quota, and there’s no quota yet to attach a rate to.

This isn’t an invented workaround. It mirrors a structure sales organizations already use when a rep lacks a track record to price commission against: the draw-against-commission model. HubSpot’s benchmarking, cited by Xactly, puts average time to full ramp at about 3.2 months for new reps generally, though other current benchmarking puts AE-specific ramp time considerably longer, often four to six months for a mid-market motion and longer still at enterprise scale. A common industry step-down structure for the draw itself looks like this: a full draw in months one through three, 75% in months four through six, and 50% in months seven through nine, tapering as production data accumulates. Either way, the shape is the same: a rep draws against guaranteed pay while their production data is still too thin to run standard commission against, then transitions onto the full plan once enough closed deals exist to measure them fairly. The adaptation for a founder-led sales org is the same logic applied one layer earlier: instead of drawing against a commission plan that will eventually kick in on a schedule, you’re drawing against a quota that will eventually exist once there’s enough closed data to build it honestly.

Stage 2 Capital, whose LPs are practicing GTM operators, runs the Catalyst program for early-stage B2B SaaS founders as part of its broader GTM curriculum. First-comp-plan design for a founder-led sales motion sits squarely inside the kind of problem that curriculum exists to address: the plan is a founder-education exercise about what your sales motion actually looks like, not a template exercise. A steady-state, quota-based plan is a template exercise once that motion is proven. Applying the second kind of thinking to the first kind of problem is where these plans go wrong.

Reintroducing Variable Pay

None of this means base-only forever. Once enough deals have closed to build a quota from actual cycle length, actual win rate, and actual segment ACV rather than a founder’s intuition, standard quota-based variable pay is appropriate again, and the rep should expect it. The threshold isn’t a fixed number of months, it’s enough closed data to make the quota a fact instead of a guess.

At Chalk Theory, this is exactly the kind of gap our core GTM/RevOps practice diagnoses before recommending a fix: comp design only works once the underlying data model, cycle stages, win/loss capture, does.