SaaS Metrics for Compensation Planning
Comp plans amplify whatever you measure. This page maps each SaaS role to the right primary KPI, sets hurdle bands using benchmark percentiles, and lays out the variable comp ratios that actually produce the behaviour the business wants.
Why most SaaS comp plans break
The single most common comp-plan failure mode in SaaS is tying variable pay to a metric that is easy to count but does not align the role to outcome. CS comp tied to gross renewals incentivises holding accounts that should have been churned for fit. Marketing comp tied to MQL volume incentivises noise instead of pipeline yield. Sales comp tied to bookings without ramp adjustment incentivises closing-the-quarter discounting that damages the forward NRR. In each case the metric is the wrong proxy for the outcome the business actually wants.
The fix is structural, not motivational. Pick the metric that aligns the role to outcome over a defensible time window. Use benchmark percentiles to set the hurdle band. Use accelerators only above target. Build clawbacks where the cycle from comp event to outcome is longer than the variable pay period.
The reference set for the bands below is the Pavilion State of Compensation report, the BLS OES wage tables for engineer, sales, and customer success base bands, and Sales Hacker compensation studies for sales accelerator structures. All benchmark numbers verified May 2026.
Metric to role mapping
| Role | Primary KPI | Secondary | Hurdle band | Comp split |
|---|---|---|---|---|
| Sales (AE) | Quota attainment (new logo ARR) | Pipeline coverage, win rate | 60-70% AE at quota, $0.8-1.5M ARR per rep at $25M+ scale | 50/50 base/variable, accelerators above 100% |
| Sales (SDR) | Qualified pipeline created | Meeting-to-opportunity conversion | 20-40 qualified meetings per month outbound; varies by ACV | 70/30, accelerators on opportunity conversion |
| Customer Success | NRR (with GRR floor) | Expansion ARR, logo retention | NRR 105-115% per CSM book, GRR 92%+ floor | 80/20 to 70/30, 12-month expansion clawback |
| Product | Activation + adoption (PLG: PQL conversion) | Feature usage breadth, NPS | Activation 40-60% of new signups for PLG; case-specific for sales-led | Company bonus pool, no individual variable |
| Marketing | Pipeline coverage and CAC payback | MQL-to-SQL, brand search lift | 3x pipeline coverage, CAC payback under 18 months | 80/20, accelerators on pipeline-to-closed-won |
| RevOps | Forecast accuracy (deviation from booked) | Conversion improvement, cycle time reduction | Forecast within 5% for quarter, 10% for following quarter | 85/15, variable on accuracy and improvement |
| Engineering (IC) | Company KPIs (ARR, Rule of 40) | Team reliability + velocity | Company bonus pool target hit | 100/0 base, with annual company bonus |
| Engineering leadership | 70/30 blend (operational + company) | Reliability, attrition, velocity | Reliability SLO met, voluntary attrition under 12% annualised | 75/25, blended scorecard |
Setting hurdle bands using benchmark percentiles
Hurdles tied to internal-only history are vulnerable to two failure modes: hurdles too easy (variable pays out at non-stretch performance) and hurdles too hard (variable never pays out and the plan becomes demotivating). Hurdles tied to external benchmarks avoid both.
The right framework is to set the hurdle at the median of the appropriate benchmark cohort and set the accelerator threshold at the top quartile of the same cohort. For a $10M ARR SaaS company, that means CS hurdle on NRR of 108 percent (median of $10M ARR cohort) and accelerator threshold of 118 percent (top quartile). The variable plan then pays target for hitting the median (the bar for being competent at the company's scale) and pays accelerated for beating the top quartile (the bar for outperformance).
Refresh the bands annually using the latest cohort benchmark data. Cohort benchmarks move (NRR medians compressed roughly 8 points from 2021 to 2026), and a hurdle band frozen at 2021 levels would be unattainable today. The matching dataset is on our benchmarks by ARR tier page.
Accelerators and decelerators
Accelerator structures vary by role and cycle time. The principles that hold across roles:
- Accelerators kick in only above 100 percent attainment. At-quota accelerators distort the plan. Above-quota accelerators reward genuine outperformance.
- Accelerator rates of 1.5x to 2x of base commission rate. Higher than 2x creates outsized rep-comp swings that distort the P&L. Lower than 1.5x is not actually motivating.
- Decelerators on the bottom 50 percent of attainment for ramped reps. A rep at 30 percent attainment in their fourth quarter post-ramp is a hiring miss. Comp plan should reflect that, with a performance-improvement plan as the right next step.
- Cap rare and high. A cap that bites in a real quarter destroys motivation. If you cap, cap at 200 percent of variable. Better still, do not cap.
- Clawback for revenue that does not stick. 12-month expansion clawback for CS, 6-month new-logo clawback for sales (refund commission on accounts that churn inside 6 months of close). This protects the comp plan from incentivising low-quality bookings.
Common comp-plan anti-patterns to remove
Paying CS on gross renewal rate only. Incentivises CSMs to fight to hold every account, including the ones that should have been churned for fit. Net retention with gross retention floor is the correct primary metric.
Paying marketing on MQL volume. Incentivises noise. Tie marketing variable to MQL-to-SQL conversion and pipeline-to-closed-won conversion. The right metric is yield, not volume.
Paying sales without ramp adjustment. New reps need 3 to 6 months to ramp depending on cycle length. Paying full variable from day one creates either bad hires collecting full base in their ramp, or rep-quality issues being masked by the lag.
Quarterly variable on a long-cycle role. Engineering team variable tied to quarterly product KPIs creates shipping pressure that damages quality. Engineering variable should be annual, tied to company KPIs.
No clawback on expansion bookings. Without clawback, CS reps can land low-quality expansion that churns inside the year. 12-month clawback aligns CS comp to NRR that actually persists.
The KPI calculators that feed comp plan design
The hurdle math behind the bands above comes from running the same calculators you use for board prep, then segmenting by territory (sales), book (CS), channel (marketing). Each of the per-metric tools below carries the cohort benchmark in the result panel.
For comp benchmarks by role and stage, see the Pavilion State of Compensation report (linked above) plus the BLS OES wage tables. For the metric distributions that anchor the hurdle bands, see our 2026 benchmarks overview and the public SaaS dashboard.