Why Your Comp Plan Is Paying for the Wrong Behaviour

Every sales behaviour problem has a compensation plan underneath it. The reps who discount to hit their number are being paid on revenue, not margin. The reps who sandbag are protecting a baseline they know will be used to set next year's quota. The reps who cherry-pick easy deals and avoid complex enterprise opportunities are responding rationally to a commission structure that treats a £50,000 deal and a £500,000 deal as ten times the same thing.

The behaviours are not the problem. They're the symptom. The comp plan is the problem.

This is uncomfortable, because it means the problem is not the sales team. It means the problem is the system the sales team operates inside — a system that leadership designed, finance approved, and RevOps is now trying to manage around with coaching programmes and pipeline reviews and performance improvement plans that address symptoms while leaving the cause entirely intact.

The incentive design principle

People do what they're paid to do. Not what you want them to do — what the comp plan pays them to do.

This sounds obvious, but organisations consistently act as though behaviour is a character issue rather than a design issue. The rep who discounts aggressively is not lacking integrity. They are optimising for the metric the comp plan rewards. Commission is paid on revenue closed. Discounting closes revenue. The mathematics are straightforward. The person responding to them is behaving exactly as the system was designed — even if the system was never designed with that outcome in mind.

Coaching does not fix a comp problem. A performance improvement plan does not fix a comp problem. A motivational SKO presentation absolutely does not fix a comp problem. The only thing that fixes a comp problem is a different comp plan — one designed with an understanding of the behaviours it creates, not just the costs it controls.

The reason organisations treat behavioural problems as character problems is that character problems are cheaper to address than design problems. You can put a rep on a PIP. You cannot put the comp plan on a PIP. But you can interrogate it. You can map the behaviours it creates. You can build the evidence that forces the conversation that leadership has been avoiding — usually because the last time someone tried to change the comp plan, it took six months, triggered three employment law reviews, and ended with the plan being 15% different in ways that nobody could explain.

The difficulty of changing the plan is not a reason to leave a broken plan in place. It is a reason to do the audit first, so that when the conversation happens, it happens with evidence rather than anecdote.

The most common unintended consequences

Most comp plan problems are not novel. They are predictable. The same four failure modes appear across industries and organisation sizes, because they are structural responses to structural incentives rather than individual departures from expected behaviour.

Sandbagging

Reps who sandbag are protecting their baseline. If this year's performance becomes next year's quota — which it typically does, because quota-setting is often nothing more than last year's actuals plus a growth expectation — the rational strategy is to not significantly outperform. A rep who closes 130% of quota in Q4 can reasonably expect to be handed a quota that is 20–30% higher next year. That is not a reward for success. That is a punishment dressed up as a target.

Sandbagging is not dishonesty. It is a rational response to a quota-setting process that punishes success by treating it as evidence that the bar was set too low. The rep who closes 97% of quota every year is not underperforming. They are optimising. The system taught them to.

End-of-quarter discounting

If the commission structure pays the same whether a deal closes at full price or at a 30% discount, reps will discount to close. The cost falls on margin. The commission is identical. The rep has no financial incentive to hold on price because the comp plan creates no financial disincentive to concede it.

The plan created the incentive. The sales team responded to it. When finance presents a margin report showing systematic discounting in Q3 and Q4, the conclusion should not be that the sales team lacks discipline. The conclusion should be that the comp plan lacks a mechanism to make margin matter to the person closing the deal.

Cherry-picking

Reps who avoid complex, long-cycle deals in favour of smaller, faster ones are doing exactly what the comp plan rewards — maximising commission per unit of time. If a £50k deal takes three months and a £500k deal takes nine months, and the commission rate is identical, the rational choice is clear. The rep who prioritises the enterprise pipeline is doing so despite the comp plan, not because of it.

Organisations that want reps to pursue enterprise deals need a comp structure that makes enterprise deals worth pursuing — through higher commission rates, accelerators on large deal sizes, or multipliers on strategic account wins. Without that structure, the ask is for reps to behave against their financial interests because leadership has decided it is strategically important. Some reps will. Most will not. The ones who do not are not selfish. They are rational.

Post-close disengagement

If commission is paid on signature and there is no clawback for early churn, reps have no financial incentive to care about customer success after the contract is signed. The deal is closed. The commission is earned. What happens next is a customer success problem, not a sales problem — not because reps are callous, but because the comp plan has communicated, unambiguously, that the relationship ends at signature.

The consequence is customers who were oversold into products that do not fit, who churn at month seven, and whose account executives have already moved on to the next quarter's pipeline. The comp plan created that incentive structure. It is rational to follow it.

Why comp plans go wrong

Comp plans are designed by finance for controllability, not by sales for motivation. The primary goal is predictable cost — a compensation structure that does not blow the headcount budget regardless of how the sales team performs. This is a legitimate requirement. It is also the wrong starting point for a plan designed to drive specific behaviours.

When controllability is the design constraint, you get plans that pay on closed revenue because revenue is easy to measure and attribute. You get flat commission rates because variable rates create modelling complexity. You get annual plans with no mid-year adjustment because changing the plan mid-cycle is operationally painful. All of these design choices are rational from a finance perspective. All of them create incentive problems that RevOps then spends the year trying to manage.

Comp plans are also built once and rarely reviewed. They are designed at company founding or at a major restructuring, and they persist because changing them is politically and contractually complicated. The market changes. The product changes. The go-to-market motion changes. The average deal size shifts. The sales cycle lengthens. And the comp plan stays the same, because revisiting it means admitting it was wrong, which means someone has to own that it was wrong, which means a conversation that everyone would prefer to have next year.

By the time a comp plan is visibly broken — by the time the sandbagging is undeniable and the discounting is systemic and the attainment distribution has collapsed — it has typically been quietly wrong for two or three years. The signal was always there. Nobody was looking for it.

The metric selection problem

The metric you pay on defines the behaviour you get. This is not a metaphor. It is a direct causal relationship between the incentive structure and the decisions individual reps make every day across every deal in their pipeline.

Paying on revenue rewards revenue maximisation, regardless of margin, contract length, or customer fit. Reps will close the deal that closes fastest, at the terms required to close it, because that is what the metric rewards.

Paying on ARR rewards ARR, regardless of whether that ARR retains. A rep paid on new ARR has no incentive to qualify out a customer who will churn in six months. The ARR is real at signature. The churn is a CS problem at month seven. The metric makes these two things entirely separate.

Paying on new logo acquisition rewards new logos, regardless of expansion opportunity in the existing base or churn in the installed base. Organisations that wonder why their net revenue retention is poor while their new logo acquisition is strong have answered the question themselves — they have built a comp plan that tells reps where to spend their time, and the reps have listened.

Every metric creates unintended incentives. The question is whether you have mapped them before you launch the plan or discovered them afterwards. Most organisations discover them afterwards, at considerable cost, and then treat the discovery as a surprise rather than as a predictable consequence of the design choices that were made.

How to detect sandbagging in the data

Sandbagging leaves a signature in the data. The signature is consistent enough that if you are not seeing it in your attainment reports, the more likely explanation is that you are not looking for it, not that it does not exist.

Look for reps who consistently achieve 95–105% of quota — close enough to demonstrate performance, never enough to trigger a quota increase. One year at 97% is a data point. Three consecutive years at 94–99% is a pattern. That pattern does not happen by accident in a business where deal flow is noisy and market conditions vary. It happens because someone is managing the outcome.

Look for deal timing patterns where opportunities that could plausibly close in one quarter consistently close in the next. Pull the closed-won data by quarter and look at when in the quarter deals tend to close. A healthy close distribution is roughly continuous through the quarter. A sandbagging distribution clusters in the first two weeks of the following quarter — just after the quarter closed, when the previous number was already secured.

Look for pipeline that appears in the final weeks of a quarter after the rep has already secured their number for that period. Late-appearing deals that close in the subsequent quarter are a classic mechanism for shifting revenue across a boundary without it being visible in any single quarter's pipeline review.

These patterns are visible if you are looking for them. The vast majority of organisations are not. They are running pipeline reviews to understand what will close this quarter. They are not running the data forensics that would reveal whether what is closing this quarter is being managed to a behavioural outcome rather than to a genuine sales result.

Clawback provisions — do they work?

Clawbacks exist to align rep incentives with customer success. If a customer churns within 90 days of contract signature, the rep loses the commission. The theory is sound: if reps share the downside risk of a bad fit, they will qualify more carefully and oversell less aggressively.

The evidence on whether clawbacks actually change behaviour is mixed. Reps dislike them, which creates resentment that can itself damage the customer relationship. They are difficult to enforce cleanly, particularly where multiple reps have touched an account or where churn is attributable to factors outside the rep's control — product gaps, CS failures, market shifts. The result is that clawbacks are often more symbolic than functional, applied inconsistently and contested frequently.

A well-designed clawback provision on a poorly designed comp plan is still a poorly designed comp plan. The clawback addresses one symptom — post-close disengagement — while leaving the underlying incentive structure intact. The design of the primary commission mechanics matters more than any provision added to mitigate the problems that design creates.

What a comp audit looks like

Before redesigning the plan, audit the current one. The audit produces evidence. The evidence makes the redesign conversation possible — not because leadership needs to be persuaded that a problem exists, but because the conversation about what to change requires data rather than instinct.

Pull deal margin distribution by rep. What is the average discount rate across the team? Is it consistent across reps, or do specific individuals discount systematically more than others? A rep whose average discount is 28% when the team average is 14% is responding to something. Whether that something is territory, quota pressure, product-market fit, or a comp structure that incentivises close over price is the question the data should answer.

Pull deal timing by quarter. Are closes clustering at quarter-end in ways that suggest managed behaviour rather than genuine close timing? Are deals that appear in week ten of a quarter consistently closing in week one of the next? Map the distribution and compare it to what a random close distribution would look like in your sales cycle. The deviation from random is the signal.

Pull commission payouts by deal type. Are reps being financially rewarded for the deals the business actually wants — the strategic accounts, the multi-year contracts, the high-margin enterprise wins — or are they being rewarded in proportion to volume, regardless of deal quality? If the highest earners on the commission report are not the same people closing the most strategically valuable deals, the comp plan is answering a different question than the one you think you asked.

The audit does not redesign the plan. It maps the current plan's actual outputs against its intended outputs. The gap between those two things is the evidence that the conversation requires.

The sales and finance relationship

Comp design requires both sales leadership and finance in the room at the same time. This does not reliably happen, which is one of the reasons comp plans fail so predictably.

Finance brings cost modelling and controllability requirements. Without finance, you get comp plans that are motivationally interesting and financially irresponsible — plans that accelerate aggressively without cap structures, or that create commission leverage that blows the headcount budget in a strong sales year.

Sales brings behavioural knowledge — the understanding of what reps actually do in response to specific incentives, which deals get prioritised, how quota pressure manifests in the pipeline, what the conversations in the field actually sound like when reps are deciding where to spend their time. Without sales, you get plans that are financially controlled and behaviourally counterproductive — plans that optimise for cost predictability at the expense of the incentive alignment they were supposed to create.

The conversation between these two functions is almost always uncomfortable, because finance tends to frame comp as a cost to be controlled and sales tends to frame comp as a motivation tool to be maximised. Both framings are partial. The right framing is that comp is a design problem — a mechanism for translating business strategy into individual daily decisions across the sales organisation. That framing requires both perspectives to be in the room, both sets of constraints to be on the table, and both functions to stay in the conversation long enough to reach a design that is financially responsible and behaviourally coherent.

The conversation is always uncomfortable. That is why it does not happen often enough. And that is why the same four comp plan failures appear, in the same organisations, in the same ways, year after year — not because the problems are difficult to understand, but because the conversation required to fix them is difficult to have.

The behaviours your team is exhibiting are not mysteries. They are the predictable outputs of the plan you built. The question is whether you are willing to audit what that plan is actually paying for — and act on what the data shows.