KEY TAKEAWAYS
- A comp plan only works when a rep can do the math on it in under a minute. Most plans fail not because the incentives are wrong, but because reps can’t calculate their own payout fast enough to let it guide daily decisions.
- Every metric in a comp plan should pass a controllability test. A rep needs to move the number through their own effort, see progress in real time, and connect that number to a behavior the business actually needs — otherwise the metric is just noise.
- Two to three metrics is the ceiling for a plan that changes behavior. Add a fourth or fifth to “cover all the bases” and reps will default to whichever one pays the most or is easiest to hit, making the rest irrelevant.
- Comp plans need a 90-day diagnostic, not an annual review. Quota fairness, payout skew, and accelerator abuse all show up quarterly — waiting a year to check means small misalignments turn into expensive ones.
Your sales team has a comp plan. They get paid when they hit their numbers. And yet your reps are still sandbagging deals, chasing the wrong customers, and quietly ignoring the strategic priorities you actually need them focused on.
I’ve sat across from finance teams who built a gorgeous, six-metric comp plan and couldn’t explain what it would actually do to seller behavior.
That’s the problem.
We design comp plans like tax code and then act surprised when reps treat them like one by finding every loophole while ignoring everything they can’t quickly calculate.
If compensation drives behavior, why isn’t yours driving the right behavior?
Here’s what I’ve learned, plan after plan: variable comp isn’t a payout formula. It’s a behavior design system. Every metric, threshold, and accelerator sends a signal about what matters most. When those signals conflict with what the business actually needs, reps do exactly what you incentivized — not what you wanted.
Let’s explore reasons why your variable comp plans aren’t measuring up.
Why most variable comp plans fail to change anything
Every company has a comp plan. Few can draw a straight line from that plan’s structure to what reps actually do day to day.
Gartner research found that only 24% of sales reps can easily calculate their own commissions. A plan a rep can’t do the math on isn’t a helpful tool. Moreover, they may check it once a quarter or ignore entirely.
Most plans break down in three predictable ways.
The plan rewards the wrong things. Revenue-only commissions sound simple, but they reward speed over quality. Reps discount aggressively to close faster. They chase high-velocity deals and skip past strategic accounts that take longer to develop. The plan says “maximize revenue.” The business needs sustainable growth. Those aren’t the same instruction.
The plan is too complex to act on. Companies stack metric after metric trying to cover every base — pipeline creation, deal size, margin, satisfaction, territory penetration, all weighted and interconnected. Reps default to whichever one is easiest to hit or pays the most. Everything else becomes decorative.
The plan never gets recalibrated once it’s live. Most companies design comp annually and review it quarterly. That cadence made sense when markets moved slowly. They don’t move slowly anymore. GTM priorities shift faster than an annual planning cycle can keep up with.
Sales Performance Management Software: The Complete Guide to Driving Revenue Efficiency
Here’s the deeper issue: most compensation systems work forward from a revenue number instead of backward from a behavior. In other words, what they should start with is “we need reps prioritizing multi-year deals in mid-market accounts” and build the incentives that create that motion specifically.
The disconnect between pay and performance
You’re paying on outcomes reps can’t control. The controllability principle should govern every metric in your plan: if a rep can’t move a number through their own effort within the measurement period, paying on it creates noise, not signal.
I’ve talked with numerous sales leaders who inherited a plan that paid account managers on company-wide NPS — not their book of business, the whole company’s score. They watched genuinely strong AMs get penalized for a support ticket backlog they had zero authority over.
You’re paying on one thing but asking for another. The plan says “revenue.” Leadership wants multi-year deals with strategic accounts. Those pull reps in opposite directions. When revenue is the only lever, reps optimize for speed and size — discounting to close faster, favoring transactional deals over strategic ones, ignoring expansion revenue because new logos pay a higher rate. Reps hit their number. The business absorbs the churn, the discounting, and the poor-fit customers that number left behind.
You’ve added too many metrics trying to fix the first two problems. Plans with five or six metrics meant to “cover all the bases” end up covering none of them. When people face too many priorities, they effectively act on none — a plan paying 60% on bookings, 20% on pipeline, 10% on satisfaction, and 10% on territory penetration behaves, in practice, like a bookings-only plan.
“Accelerators, decelerators, micro-incentives. I mean, remind me again, why would I do this for another 2%?” Jason Lemkin, Founder of SaaStr, candidly said about designing comp plans. ” And why are you adding decelerators that ‘punish’ activity that still adds ARR? I couldn’t understand it. And if I couldn’t get my arms around the plan … how could I champion it?”
If a rep can’t explain their plan and estimate a payout in under sixty seconds, the plan is too complex to guide a single daily decision.
Behavioral side effects you didn’t design for
Every comp structure creates behaviors you intended and behaviors you didn’t. Map both before the plan goes live.
Revenue-only commissions close deals, but they also invite heavy discounting and total indifference to retention. Uncapped accelerators without quality gates motivate overachievement, but they also produce sandbagging — reps holding deals back early in the period, then pulling future deals forward to hit a threshold.
Annual quotas with no interim checkpoints support full-year planning, but they also produce the hockey-stick Q4 rush and the rep who quietly checks out in Q3 after falling behind. SPIFs stacked on top of a base plan create short bursts of focus, but they distract from the metrics that actually drive sustained performance. Team bonuses without individual accountability sound collaborative, but they let low performers ride on the work of the team’s best sellers and your best sellers notice.
None of these patterns are surprising once you’ve seen them a few times. The fix isn’t abandoning these structures. It’s building guardrails that keep the intended behavior while limiting the side effect.
The controllability audit
Got 60 seconds? Take that time to run every metric in your plan through four questions:
- Can a single rep move this number through their own effort?
- Can they influence it within the measurement period?
- Can they see their progress in real time or close to it?
- Does this metric connect to a behavior your GTM strategy actually needs?
A “no” on any of these means the metric needs restructuring or removal.
Take a common AE plan built on bookings, multi-year deal mix, and pipeline generation. Bookings clears all four bars — reps control deal progression, can influence it within a quarter, see it live in the CRM, and it maps directly to revenue growth. Multi-year deal mix often fails on controllability if pricing and terms are set above the rep’s pay grade; the fix is measuring what the rep actually controls, like multi-year proposals presented, or dropping the metric. Pipeline generation clears the bar when it’s built around prospecting activity reps own directly, measured on a cycle they can influence, with visibility they can check daily.
That audit forces precision about what you’re actually measuring and why. Metrics that survive become behavior drivers. The ones that don’t are just distractions with a formula attached.
Building the plan by role
SDRs should be paid on the inputs that predict downstream revenue, not volume metrics that are easy to game. Weight compensation toward qualified opportunities accepted by AEs, with a smaller share tied to activity pace — typically a 60/40 or 70/30 base-to-variable split, reflecting shorter cycles and less deal control. Reward opportunities created and qualified meetings booked. Skip paying on raw call or email volume; it rewards noise, not pipeline quality.
Sales Performance Analytics: The Complete Guide to Revenue Intelligence That Drives Action
AEs should be weighted heavily toward bookings with at least one deal-quality modifier — contract length, margin, strategic segment mix, or customer fit. Companies with more than 30% variable pay for this role see meaningfully higher win rates, which supports a 50/50 base-to-variable split for AEs specifically. Accelerators above quota should matter — 1.5x to 2x the standard rate past 100% attainment. A small accelerator doesn’t move behavior; a large one creates real urgency.
Account managers and CSMs need plans that reward keeping and growing existing customers, not upselling everything in sight. Pay on net revenue retention or expansion ARR, with a floor tied to gross retention, so growth pursuits never come at the expense of keeping the account healthy. Alexander Group data shows 53% of companies now put CSMs on incentive plans tied to sales outcomes — a reflection of how directly customer success now touches revenue retention. A 60/40 or 70/30 base-to-variable split fits the longer, more consultative cycle these roles operate on.
Overlay and specialist roles — solutions engineers, industry specialists — need plans that credit their contribution without double-paying the same revenue twice. Use influenced-revenue crediting with caps or team-based components, and cap the metric count at two. Overlay plans that try to mirror a full AE plan create confusion and internal competition instead of clarity.
Stress-testing the plan before launch
Run last year’s actual deal data through the new plan and see what your team would have earned. Look hard at both ends — significant overpayment scenarios and significant underpayment ones.
Model three company-wide attainment scenarios: 80% of reps at quota, 50% at quota, 120% at quota. Confirm the company can afford the payout at the high end and that the plan still motivates at the low end.
Pilot with a small group before rolling the plan out company-wide. A subset of reps will surface unintended incentives while the mistake is still cheap to fix.
Then run the simplicity test: hand the plan to three to five reps and ask them to calculate their own payout on a recent deal. If they can’t do it quickly and get it right, the plan needs to be simpler, not better explained.
The quarterly comp diagnostic
Annual design cycles are fine. Annual review cycles are too slow. Run this diagnostic every 90 days.
Most sales leaders review comp plans the way most companies still do — once a year, well after the damage was done. But some of the most effective sales leaders ask one question every quarter: what percentage of reps are landing between 80% and 120% of quota? If that number’s off, no amount of comp tweaking fixes what’s actually a quota or territory design problem.
From there: check whether payouts are skewed toward one metric while the others sit unused — anything contributing less than 10% of total payout isn’t shaping behavior at all. Look for accelerators paid out on deals that don’t match strategic priorities; heavy discounting or poor-fit customers earning top payouts is a sign of misaligned incentives, not strong selling.
Watch attrition and exit interviews for comp complaints, which usually point to quota fairness or plan complexity rather than pay level itself. And notice when finance and sales start arguing about payout math — that argument is a symptom of a plan that’s either too complex or built on mismatched data definitions between systems.
Monthly tracking of pipeline creation, deal velocity, and average discount rates gives you the early warning quarterly reviews miss.
Rolling out changes without losing trust
Comp changes trigger an emotional reaction even when they benefit the rep. Bad communication can undo good plan design entirely.
Give reps at least 30 days’ notice before a new period starts, so they can adjust their pipeline with the new plan in mind.
Explain the business reasoning before the mechanics — reps accept a plan change far more readily when they understand why it’s happening, not just what changed.
Walk through a concrete example at 100% attainment using real numbers; weightings and percentages on a slide don’t help anyone picture their actual paycheck.
Put the plan in writing and require a signed acknowledgment — verbal explanations get misremembered and turn into disputes months later. And never touch an active measurement period; reps built their pipeline strategy around the plan that existed when they started working those deals.
Frequently asked questions
What percentage of total compensation should be variable for sales roles? Variable pay typically runs 40–50% of on-target earnings, with meaningful variation by role. SDRs often sit at 30–40% variable, AEs commonly land at 50%, and senior enterprise reps can run 60% or higher.
How often should companies review and update compensation plans? Leading companies now pair an annual design cycle with quarterly diagnostics. Monthly tracking of attainment distribution and payout patterns catches problems long before the quarterly review would.
What’s the biggest mistake companies make in compensation plan design? Paying on outcomes reps can’t control within the measurement period. A metric a rep can’t move through their own effort doesn’t motivate — it just adds confusion to the plan.
How many metrics should a sales compensation plan include? Two to three is the practical ceiling. Beyond that, reps gravitate toward whichever metric pays the most or is easiest to hit, and the rest stop influencing behavior at all.
Should compensation plans include team-based components? They can work well for overlay and collaborative roles, but only alongside individual accountability. Pure team payouts tend to let low performers coast while your strongest sellers quietly resent carrying the difference.