SaaS Sales Commission Structures: OTE Splits That Actually Work

A SaaS commission structure that works starts with the pay mix, and the market has converged. Closing AEs run a 50/50 base-to-variable split. SDRs run 70/30. Sales leaders sit at 60/40. Commission lands between 8% and 12% of annual contract value, with the B2B SaaS median at 11.5% at full quota, and quota is set at 4x to 6x OTE. That is the easy part. The hard part is the quota sitting underneath the split. Only 48% of reps hit quota in 2026. A plan priced for 100% attainment is a plan that pays out at 60%, and your candidates worked that out before they signed.

Key takeaways

• 50/50 is the AE standard and the median across B2B SaaS now sits at 53/47. SDRs run 70/30, first-line managers 60/40, VPs and CROs 60/40 to 70/30.
• Commission rates cluster at 8-12% of ACV, median 11.5% at 100% of quota. Quota multiples have held at 4x to 6x OTE since 2022.
• Only 48% of reps hit quota in 2026, down from 51% in 2024. More than 58% of companies over-assign quota by 20-30% to cover for it.
• 53% of SaaS companies run clawbacks, usually on a 90 to 120 day churn window. Narrow windows work. Broad ones cost you your best reps.
• Pay mix is not portable. A 50/50 plan that reads as confident in San Francisco reads as risk-shifting in Singapore, and your offer acceptance rate will say so.

What is the right OTE split for each sales role?

The split should track how much of the outcome the rep actually controls. A closing AE controls the close, so half their pay can ride on it. An SDR controls a meeting, not a deal, so most of their pay should be guaranteed. A CRO controls a system that takes three quarters to move, so their variable should be measured over quarters, not months.

Here is where the market sits in 2026 across the roles we place most often:

Two things to notice. First, the quota multiple falls as deal size rises. Enterprise AEs carry three to four times OTE because cycles are long and variance is brutal, and pretending otherwise produces a plan nobody can model. Second, the payout cadence changes with the sales cycle. Pay an enterprise AE monthly on a nine-month cycle and you have built a lottery with a payroll department attached.

OTE bands themselves vary more than the splits do. SMB AEs land at $110K-$150K, mid-market at $140K-$200K, and enterprise AEs working six-figure deals at $220K-$320K. The average across all AE segments is around $154K. For the leadership layer above them, our VP of Sales salary guide covers the same markets in detail.

How much commission should you pay per closed deal?

Between 8% and 12% of ACV, and the exact number is not a choice you get to make freely. Three inputs lock together: OTE, quota, and commission rate. Set any two and the third is already decided.

Run the arithmetic. A mid-market AE on $160K OTE at a 50/50 split has $80K of variable. Put them on a 5x quota multiple and they carry $800K of new ARR. The commission rate that pays $80K on $800K is exactly 10%. Change the quota to $650K and the rate has to rise to 12.3% or the rep earns below OTE at full attainment. Change the rate to 8% and quota has to fall to $1M, which no longer matches a 5x multiple.

This sounds obvious written down. In practice we watch companies set OTE from a competitor's job ad, set quota from the board deck, and set the commission rate from a blog post, then discover in month four that a rep at 100% of quota earns 78% of the number on their offer letter. That rep does not raise a ticket. They take a call from a recruiter.

So build the plan in that order: decide what a good rep should earn, decide what a good rep can realistically sell, then derive the rate. If the derived rate is above 13% of ACV, your quota is too low or your OTE is too high for the deal size you sell. If it is below 7%, your quota is fiction.

Why do most commission plans break in the second year?

Because year one is priced on hope and year two is priced on a growth target. The plan does not fail because the maths was wrong at launch. It fails because the maths gets rewritten every January while the sales motion stays the same.

The pattern repeats across the 555+ GTM placements we have made in 35+ countries. The board asks for 2.5x growth. Headcount rises by 40%. The difference is closed by raising per-rep quota by 30%, which is exactly the over-assignment that 58% of companies now build into their plans. Attainment drops. The reps who were carrying the number leave first, because they are the ones with options. The remaining team misses harder, and next January the quota rises again to compensate for the miss.

There is a version of this that works, and it is unglamorous. Set quota from your own conversion data rather than from the number the plan needs to be true. Model the payout at your actual historic attainment distribution, not at 100%. If your median rep closes $600K and your plan only funds itself at $850K, you do not have a comp problem, you have a pipeline problem wearing a comp costume. Fixing the plan will not fix it.

The second protection is a written change policy. Tell reps at the start of the year what can and cannot be altered mid-year: territory, quota, accounts, rate. Companies that reshuffle territories in Q3 and act surprised by Q4 attrition are paying a retention cost they never book anywhere.

Do accelerators and clawbacks actually work?

Accelerators do, if they are uncapped. Clawbacks do, if they are narrow. Most companies get both backwards.

On accelerators, the structure that holds up is simple: standard rate to 100% of quota, then 1.5x the rate from 100% to 150%, and 2x above 150%. Uncapped. The instinct to cap kicks in the first time a rep lands a whale and earns more than their manager. Resist it. In a market where 48% of reps hit quota at all, the handful who blow through it are subsidising the entire team, and the money you save by capping them is the cheapest possible way to lose your top two performers. We have seen a capped plan cost a client a rep carrying 40% of regional pipeline over a $28K clawback on one deal.

Decelerators below quota are the other half, and they are less popular than they should be. Paying 60-70% of the standard rate below 80% of quota, then full rate above it, funds the accelerators without changing plan cost. It also makes the threshold real. A plan where 40% attainment and 90% attainment pay nearly the same rate is not a commission plan, it is a variable-rate salary.

On clawbacks, 53% of SaaS companies now run them, and the sensible window is 90 to 120 days. That covers the failure mode clawbacks exist for: a rep closing a customer who was never going to stay. Stretch the window to twelve months and you are asking the rep to underwrite churn caused by onboarding, product gaps or a champion changing jobs, none of which they control. Reps price that risk into their offer expectations, and the best ones simply decline. Keep the window short, apply it to non-payment and immediate churn, and put it in writing before the rep signs, not in a plan document circulated in month two.

How should multi-year and usage-based deals be paid?

This is where most plans quietly leak money. Paying full commission on total contract value for a three-year deal at signature feels like a growth incentive. It is a cash flow problem with a bonus attached, and if the customer churns in year one you have paid three years of commission on one year of revenue.

The structure that works: pay full rate on first-year ACV, then a reduced rate of 3-5% on years two and three when those years actually invoice. The rep still has a reason to sell length. Finance still pays out against cash that exists. For heavily discounted multi-year deals, apply a modest rate reduction rather than a flat ban, because a flat ban just moves the discount into a side letter nobody models.

Usage-based revenue needs its own answer. Commissioning on committed contract value alone rewards a rep for a large commitment the customer never consumes. Commissioning on consumed revenue makes the rep wait two quarters to get paid. The workable middle is to pay the standard rate on the committed floor at signature and a smaller expansion rate on consumption above it, measured quarterly. It costs more to administer. It is also the only version that does not systematically reward the wrong behaviour.

How does commission design change outside the US?

More than most US headquarters expect, and importing the home plan unchanged is the single most common mistake we correct. Having placed GTM talent in 35+ countries across 48 unicorns, three patterns hold.

Singapore is base-heavy. Employment Pass holders carry visa dependency, and the risk calculus shifts hard toward guaranteed pay. A 50/50 AE plan that closes candidates in Austin loses them in Singapore. Budget 60/40 and expect the negotiation to be about base rather than upside. The upside argument lands only once the candidate believes the base covers their downside.

Sydney compresses variable further, commonly 60/40 or 65/35, with superannuation sitting on top of every number you quote. Companies that price Sydney roles off a US band and forget super come in roughly 12% light before the conversation starts.

Japan and Korea are the sharpest break. Aggressive variable pay is not read as an opportunity, it is read as instability. Plans that work run 70/30 or 75/25 with a team component, and the accelerator conversation is better handled through bonus structures than raw commission. Our work building the APAC bench for Forter, which went from 0 to $30M ARR in the region in 2.5 years, ran on regionally priced plans rather than a single global template.

The same logic applies further down the org. If you are setting SDR pay in these markets at the same time, our SDR compensation benchmarks cover the four markets side by side, and our guide to founding AE compensation covers the first closer specifically, where the standard splits mostly do not apply.

FAQ