How Do I Keep Reps From Gaming the Comp Plan?

I have what I need — the task prompt's 13 hard requirements are self-contained. Here is the rewritten essay body:

---

Why Reps Game the Comp Plan in the First Place

Before you can stop gaming, you have to accept that reps are behaving *rationally*, not maliciously. A comp plan is a set of instructions. If the instructions say "we will pay you the most for raw bookings," a smart rep optimizes for raw bookings and ignores everything the plan does not pay for. This is not a character flaw — it is exactly what a well-designed incentive is supposed to produce. The problem is that the incentive was pointed at a proxy instead of the real goal.

Economists have a name for this. Goodhart's Law states that "when a measure becomes a target, it ceases to be a good measure." The instant you attach real money to a number, people manage the number, not the underlying reality it was meant to represent. Bookings were a fine health signal until you paid on them; then reps learned to inflate, time-shift, and discount their way to bookings that look great and mean little.

There are three structural reasons gaming shows up:

• Single-lever plans. If 90% of the payout rides on one metric, that metric becomes the entire game. Diversifying what you pay on is the single highest-leverage change most teams can make.
• Threshold cliffs. Plans with hard gates — "you earn nothing until 80% of quota, then everything unlocks" — create powerful incentives to *pull* deals across the line early or *push* them past it late (sandbagging). Cliffs manufacture the very timing games you are trying to prevent.
• Slow feedback. When plans are opaque and reps only learn their true payout at quarter-end, disputes and manipulation flourish in the gap. Transparency removes the shadows where gaming lives.

The takeaway: gaming is a *design* outcome, not a *people* problem. You will never coach or threaten your way out of a plan that pays more for the game than for the job. You have to change what the plan rewards.

Diagnose the Gaming Before You Redesign

Do not rewrite the plan on instinct. Spend two to three weeks pulling the data and naming the *specific* behaviors happening on your team, because the right fix depends on the trick. Below are the most common gaming tactics, how they show up in the data, and the mechanism that produces each one.

Sandbagging. Reps hold closed-won deals and slide them into the next period to smooth attainment or bank against a future accelerator. Signal: clusters of deals closing in the first week of a new quarter that were "committed" the prior quarter; unusually lumpy close dates around period boundaries. Mechanism: cliffs and accelerators reward timing over truth.

Pipeline stuffing. Reps create low-quality opportunities to hit activity or pipeline-coverage targets. Signal: high opportunity-creation volume with low stage-2 conversion; many deals sitting in the first stage with no next step. Mechanism: you paid or measured on pipeline *quantity* rather than *quality*.

Discount-to-close. Reps trade margin for a faster, easier close because they are paid on revenue, not profit. Signal: average selling price drifting down; a spike in max-discount deals near period-end. Mechanism: the plan is blind to gross margin.

Renewal dumping / churn-and-earn. In plans that pay full commission on new logos but little on retention, reps chase logos they suspect will churn, bank the commission, and move on. Signal: high first-year churn concentrated in specific reps' books. Mechanism: no clawback and no retention weighting.

Forecast manipulation. Reps "commit" deals to look sharp, then miss — or hide real deals to beat a lowered number later. Signal: chronic gap between committed and closed; conversation data that contradicts the CRM stage.

Product cherry-picking. Reps push only the fast, easy SKU and ignore the strategic product the business actually needs to grow. Signal: revenue concentration in one line; near-zero attach on the strategic add-on.

Rank these by dollar impact before you touch the plan. If discount leakage is costing you six points of margin and sandbagging costs you nothing but a bumpy chart, fix the discount problem first.

Build a Weighted, Multi-KPI Scorecard

The core anti-gaming move is to stop paying for a proxy and start scoring the whole job. Build a scorecard with these steps:

Step 1 — List every KPI a complete rep produces. For a typical B2B SaaS AE that is usually six to eight lines: new bookings (ARR or TCV), gross margin or discount discipline, expansion/attach, gross or net retention, forecast accuracy, and a quality signal like sales-accepted-opportunity conversion. If a behavior matters and is not on the card, it is invisible — and invisible is exactly where gaming hides.

Step 2 — Weight what matters, and keep the count honest. Assign each KPI a weight that sums to 100%. A common healthy split is roughly 50–60% on the primary revenue metric, 15–20% on margin/discount discipline, 10–15% on retention or expansion, and 10% on forecast/hygiene. The rule of thumb from sales-comp practice: no more than three heavily-weighted components. Beyond three, reps cannot mentally optimize the plan and it stops driving behavior — you get complexity without control. Use the extra KPIs as *gates or modifiers*, not as more headline weights.

Step 3 — Score each rep on a 1-to-5 level per line. Level 5 is elite, level 3 is meets-expectations, level 1 is a problem. The composite is simply the sum of *(weight × level)* across all KPIs. A rep who is a level 5 on bookings but a level 1 on margin, forecast, and retention lands a mediocre composite — the card makes the one-trick pony impossible to hide.

Step 4 — Decide where the composite bites. You have two options. The lighter-touch version keeps commission on the primary metric but uses the composite for coaching, ranking, promotion, and territory assignment. The stronger version wires part of variable pay — often a 10–25% modifier — directly to the composite so the paycheck itself follows the whole job. Most teams start with the first and graduate to the second once the scorecard is trusted.

Step 5 — Publish it. Every rep sees their levels and their composite, updated at least monthly. Transparency is not a nicety; it is the enforcement mechanism. When a rep can see that discounting to close just dropped their margin level from 4 to 2, the behavior self-corrects without a single manager conversation.

The scorecard's real superpower is *adaptability*. When you spot a new trick — everyone parking deals in one stage, or padding volume with tiny low-margin logos — you re-weight the matrix and the whole team re-aims the next day, with no lawyerly plan-document rewrite.

Wire Guardrails Into the Plan Math

A scorecard tells you *who* is playing straight. Guardrails in the commission formula make it *unprofitable* to play crooked in the first place. These are the standard mechanisms, with the trade-offs practitioners weigh:

Caps and windfall clauses. A cap limits payout above a threshold; a windfall clause carves out abnormally large deals for a lower, negotiated rate. The upside: you avoid paying a full accelerator on a lottery-ticket deal a rep half-inherited. The downside: hard caps demotivate your best closers and can push top talent out the door. The compromise most teams land on is *soft* handling — accelerators that keep paying above quota but at a declining rate, plus a case-by-case windfall review for anything over, say, 3–5× normal deal size.

Decelerators on discounting. Instead of paying flat commission on revenue, tie the rate to margin or discount depth. A deal at list might pay 10%; a deal at 40% off might pay 5%. This directly prices the discount-to-close game out of existence and aligns the rep with the CFO. It is the single most effective guardrail against margin erosion.

Clawbacks (chargebacks). If a booked deal cancels, fails to pay, or churns inside a defined window, the commission reverses. Common windows run 90 to 180 days; some subscription businesses hold a portion until first renewal. Clawbacks are the antidote to renewal dumping and churn-and-earn. Two cautions: keep the window and the math *crystal clear* in writing, and be aware that in some U.S. states clawback enforceability is legally constrained — have counsel review before you deploy one.

Gates and multipliers. Make part of the payout *conditional* on a hygiene metric. For example, a rep only unlocks the full accelerator if forecast accuracy is above 80% and required CRM fields are complete. This turns clean behavior into a prerequisite for the big money instead of an afterthought.

Smooth the curve, kill the cliffs. Replace all-or-nothing thresholds with a gradual ramp. If a rep earns *something* from the first dollar and the accelerator phases in smoothly around quota, you remove the incentive to time-shift deals across an artificial line.

A practical warning: every guardrail adds complexity, and complexity is its own failure mode. If a rep needs a spreadsheet and an afternoon to understand how they get paid, the plan will not drive behavior. Add the two or three guardrails that address your *actual* top gaming patterns from the diagnosis, and stop there.

Fix the Data and Process, Not Just the Payout

Most gaming is *data manipulation* that happens well before payout is ever calculated. If reps can freely edit close dates, invent deal stages, and self-approve discounts, no comp formula can save you. Harden the process:

• Required close-reason and stage-entry validation. Force a reason code on every closed deal and require exit criteria to advance a stage (a scheduled next step, a confirmed economic buyer, a signed order form). This kills pipeline stuffing at the source because a junk opportunity cannot advance.
• Discount-approval thresholds. Anything past a set discount (say, 15%) routes to a manager; past a deeper level (say, 30%), to a deal desk or finance. Reps stop trading margin casually when someone has to sign off.
• Deal desk review for large or unusual deals. A second set of eyes on structure, timing, and terms catches the sandbag and the windfall before they hit the ledger.
• Locked fields near period boundaries. Restricting who can change close dates in the final days of a quarter removes the mechanical ability to time-shift.
• Audit cadence. Sample deals monthly and re-verify the underlying facts against the CRM. Conversation-intelligence tools can surface the gap between what a rep *said* on a call and what the CRM claims, catching pipeline theater before it becomes a paid-out forecast miss.

Think of it as a two-layer defense: the *process* controls make gaming harder to execute, and the *comp math* makes it unprofitable even when it slips through.

A Step-by-Step Rollout Plan

Redesigning a comp plan is organizationally sensitive — reps' livelihoods are attached to it, and a clumsy rollout breeds distrust that is worse than the original gaming. Sequence it deliberately:

1. Weeks 1–2: Diagnose. Pull four to eight quarters of deal-level data. Name the specific gaming patterns and rank them by dollar impact. Interview a few honest top reps — they *know* where the loopholes are and will often tell you.
2. Weeks 3–4: Model. Draft the new scorecard weights and guardrails. Then run the plan backward against last year's actuals before you launch it. Modeling tells you which lever a rep *would* exploit and whether your best performers would earn less under the new design (if they would, you have a retention problem to solve first).
3. Week 5: Align finance and sales. Weights are a joint decision. Finance owns margin and cost-of-sales; sales owns motivation and quota fairness. If they set weights separately, the plan will contradict itself.
4. Week 6: Communicate early and honestly. Explain *why* the plan is changing in terms of fairness — "we want to reward the whole job, not one number" — not surveillance. Show every rep exactly how they would have been paid under the new plan last quarter.
5. Launch with a transition guarantee. Consider a one-quarter guarantee or a blended payout during the switchover so reps are not financially shocked while they adapt.
6. Review monthly, adjust quarterly. Watch the leading indicators (ASP, first-year churn, forecast accuracy, discount depth). Re-weight when a new pattern appears. Hold the plan's *core* stable for a full year so reps can plan their lives, but keep the weights nimble enough to close a fresh loophole overnight.

Trade-offs and Failure Modes to Avoid

Even a well-designed anti-gaming plan can backfire. Watch for these traps:

Over-complexity. The most common failure. In the effort to close every loophole, teams pile on so many KPIs, gates, and modifiers that no rep can hold the plan in their head. A plan that cannot be understood cannot motivate. Keep headline components to three, and use gates/modifiers sparingly for the rest.

Punishing honesty. If clawbacks and audits feel like a dragnet aimed at everyone, honest reps resent it and disengage. Frame guardrails as fairness — "we protect the reps who do the whole job from being out-earned by the ones who game it" — and reserve heavy scrutiny for genuine outliers.

Chasing every game to zero. Some minor gaming is cheaper to tolerate than to eliminate. If closing a loophole costs you more in plan complexity and rep goodwill than the loophole leaks, leave it and watch it. Perfect is the enemy of paid.

Changing the plan too often. Adaptable *weights* are good; a plan whose fundamental structure lurches every quarter destroys trust and makes reps hedge everything. Hold the architecture stable annually; flex only the weights within it.

Ignoring the manager layer. No formula substitutes for a manager who reviews deals, coaches on the scorecard, and enforces process. The plan sets the incentives; the manager enforces the culture. Both are required.

Confusing plan design with quota-setting. A fair plan on top of an unfair quota still drives gaming — reps facing an impossible number will sandbag and manipulate to survive. Get the quota right *and* the plan right; neither fixes the other.

The honest bottom line: you cannot make a comp plan perfectly ungameable, because any measure attached to money invites management of that measure. What you *can* do is spread the reward across the whole job, harden the data the plan runs on, keep the scorecard visible, and stay fast enough to re-weight when a new trick appears. Do that, and gaming stops paying — which is the only thing that reliably makes it stop.

FAQ

What is the most common way reps game a single-metric comp plan?

The most common games are discounting to close and sandbagging. When pay rides on revenue alone, reps trade margin for an easier, faster close — eroding profit the plan never measured. When the plan has attainment cliffs or accelerators, reps time-shift deals across period boundaries, holding closed business to bank against a future bonus or smooth their attainment. Both are rational responses to a plan that rewards one number and ignores the rest of the job.

How does a weighted multi-KPI scorecard actually stop gaming?

It removes the single lever. When you score bookings, margin, retention, expansion, and forecast accuracy — each with a weight and a 1-to-5 level — the composite reflects the whole job. A rep who maxes bookings but tanks margin and churns their book lands a mediocre composite, so juicing one number no longer produces a big payout. The math makes the trade-off visible and unprofitable, which is what changes behavior.

What is a clawback and when should I use one?

A clawback (or chargeback) reverses a rep's commission when a booked deal cancels, fails to pay, or churns inside a defined window — commonly 90 to 180 days, sometimes held until first renewal. Use one when you see reps chasing logos they suspect will churn just to bank the commission. Two cautions: write the window and math with total clarity, and have counsel review it, because clawback enforceability is legally constrained in some U.S. states.

Won't adding caps and clawbacks just demotivate my best reps?

It can, if you use hard caps and blanket clawbacks. The fix is to use *soft* mechanisms: accelerators that keep paying above quota but at a declining rate rather than a hard ceiling, windfall clauses that handle only lottery-ticket deals case-by-case, and clawbacks scoped to genuine early churn rather than every deal. Frame the guardrails as protecting honest reps from being out-earned by gamers, and your top performers usually support them.

How often should I change the comp plan?

Hold the plan's *core structure* stable for a full year so reps can plan their lives and trust the deal. But keep the *weights* nimble — when you spot a new gaming pattern, re-weight the scorecard promptly to close the loophole. The distinction matters: changing weights within a stable architecture closes loopholes without breaking trust; changing the fundamental structure every quarter makes reps hedge and destroys confidence in the plan.

Isn't gaming a sign I hired the wrong people?

Usually not. Gaming is a design outcome, not a character flaw. A comp plan is a set of instructions, and smart, motivated reps optimize exactly what you pay them to optimize — that is what a good incentive is supposed to do. If the plan rewards a gameable proxy, even your best people will manage the proxy. Fix the incentive before you question the hire; you will almost always find the plan was pointing at the wrong target.

Sources

• Harvard Business Review — "Motivating Salespeople: What Really Works": https://hbr.org/2012/07/motivating-salespeople-what-really-works
• Wikipedia — Goodhart's Law ("when a measure becomes a target, it ceases to be a good measure"): https://en.wikipedia.org/wiki/Goodhart%27s_law
• Xactly — sales compensation and incentive management resources: https://www.xactlycorp.com/
• CaptivateIQ — incentive compensation management: https://www.captivateiq.com/
• QuotaPath — commission tracking and compensation planning: https://www.quotapath.com/
• SHRM — sales incentive and commission plan guidance: https://www.shrm.org/
• Gong — revenue intelligence and forecast/deal-risk signals: https://www.gong.io/

Related on PULSE

• [How Do I Set Fair Sales Quotas Reps Will Actually Hit?](/knowledge/q214)
• [How Do I Build a Sales Compensation Plan From Scratch?](/knowledge/q1465)
• [How Do I Forecast Revenue When Reps Sandbag the Pipeline?](/knowledge/q1100)
• [How Do I Stop Margin Erosion From Rep Discounting?](/knowledge/q1106)
• [How Do I Design Accelerators That Motivate Without Overpaying?](/knowledge/q1112)