Sales compensation is the most consequential document a revenue leader writes. Every line in the plan is a behavioral instruction. Reps optimize ruthlessly for what pays, and they ignore what does not — no amount of coaching, culture, or MBOs will override a comp plan pointed at the wrong outcomes. Get the plan right and the team executes the strategy without being told. Get it wrong and the team executes the plan while quietly ignoring the strategy.
This playbook is for founders and revenue leaders designing or redesigning a sales comp plan. It walks through the six elements of a functional plan — OTE benchmarking, base/variable mix, quota setting, accelerators, SPIFFs and clawbacks, and the review cadence — and the failure patterns that produce comp drama instead of predictable revenue.
On-Target Earnings is where the plan starts. Under-market OTE attracts under-market reps, and no amount of clever accelerator design fixes a base compensation number that is not competitive. Benchmark by role, region, deal size, and vertical using at least three data sources (Bridge Group, Pave, RepVue, and your recruiter's market data all work), and pay at or slightly above the 50th percentile for the roles you want to win.
The common mistake is anchoring to what you paid last year. Sales markets move faster than most compensation review cycles, and reps talk. If your OTE has drifted 15% below market, you will lose your top performers within two quarters — regardless of how much they love the product.
The base/variable split signals what kind of rep you want. The functional benchmarks:
SDR / BDR: 70/30 base-to-variable. The role is upstream of revenue and needs base predictability.
Mid-market AE: 50/50. Enough base to survive a bad quarter, enough variable to reward performance.
Enterprise AE: 50/50 to 60/40 base-heavy. Long cycles make pure commission plans unsustainable.
Sales Engineer: 75/25 or 80/20 base-heavy. Technical talent expects predictability.
Customer Success Manager: 70/30 to 80/20 base-heavy. See the CS playbook — CSMs should not be paid like AEs.
The mistake founders make is pushing variable higher than these benchmarks under the theory that "more variable = more motivation." What actually happens is the best candidates take offers from competitors with more predictable comp, and the reps who accept the aggressive mix churn out during their first slow quarter.
Quota is the biggest single lever in the plan and the one most often set by feel. The functional approach:
Quota should be 4-6x OTE for a healthy plan. Below 4x, the unit economics do not work for the company. Above 6x, the quota is unattainable and reps will disengage.
Aim for 60-70% of the team hitting quota. If 100% hit, the quota is too low. If 30% hit, the quota is too high and you will lose your middle performers, who are the ones actually keeping the business running.
Set quotas based on territory potential, not on last year's performance. Punishing high performers by raising their quotas fastest is the surest way to lose them.
Communicate quota changes 60+ days before they take effect. Mid-year quota resets destroy trust faster than almost anything else a revenue leader can do.
Accelerators are the incentive structure above quota. Design them to reward overperformance without breaking the P&L:
A typical structure: 100% commission rate up to 100% of quota, 150% rate from 100-150%, 200% rate above 150%. This means a rep at 150% of quota earns 175% of their variable comp (0.5 × 100% + 0.5 × 150%). This math should be memorized by every rep — if they cannot recite it, you have not communicated it enough.
Cap accelerators only if you must. Uncapped plans attract top performers; capped plans send them elsewhere. If you cap, cap high (300%+ of quota) and communicate it in the plan document, not in a surprise conversation after a rep closes a whale.
SPIFFs (short-term performance incentives) can be useful for driving specific behavior in a limited window — a new product launch, a Q4 push, a strategic account program. They should be:
Capped at 10-15% of quarterly variable, so they do not distort the base plan.
Announced in advance with clear rules. Retroactive SPIFFs create expectations that will haunt you for years.
Clawbacks (recovering commissions on deals that churn or fail to pay) are essential and controversial. The functional pattern: 100% clawback if the customer cancels or fails to pay within 90 days of close, sliding to 0% by 12 months. This protects the company from bad-deal incentives without punishing reps for churn they could not predict.
Comp plans need to evolve as the business evolves. The cadence that works:
Annual plan reset. The full plan is redesigned once a year, communicated 60+ days before it takes effect, and locked for the year. No mid-year quota resets.
Quarterly component review. SPIFFs, contests, and non-core plan elements are reviewed quarterly. Core comp (OTE, base/variable, quota, accelerators) is not touched.
Monthly attainment review. How the team is pacing, who is at risk of missing, who is over. This drives coaching, not plan changes.
Companies that change core comp mid-year lose reps. Companies that never change comp lose competitive position. The annual reset with mid-cycle SPIFF flexibility is the balance that works.
The over-engineered plan. The plan has 12 metrics, 4 gates, and 3 modifiers. Reps cannot calculate their own paycheck. Fix: reduce to 2-3 metrics maximum, printable on one page.
The MBO trap. A large portion of variable comp is tied to subjective objectives (MBOs). Reps game the objectives and ignore the revenue. Fix: MBOs are fine at 10-15% of variable, never more. Everything above that is revenue-linked.
The comp-driven strategy. The plan drives behavior the company does not actually want (over-discounting to close, front-loading deals to hit quota, ignoring renewals). Cause: the plan was not stress-tested against edge cases. Fix: run scenarios before finalizing the plan.
The retroactive change. Leadership realizes the plan is over-paying and changes terms mid-year. The top performers leave within a quarter. Fix: honor the plan even when it costs the company, and redesign for next year.
The pay disparity blindness. Reps in the same role earn wildly different amounts because of territory differences that are not adjusted for. Fix: territory scoring and quota normalization, published to the team.
A well-designed comp plan is invisible — reps focus on selling, not on their pay math, because the plan rewards the exact behavior the company needs. A poorly designed plan is a constant source of drama, negotiation, and turnover. The difference is not in how clever the plan is; it is in how honestly the plan reflects what the company actually wants sales to do.
