Compensation plans are frequently designed by finance for cost control and treated by sales as a target. They are actually a behaviour specification, and they work.
The basic structure
Base salary providing stability, and variable compensation tied to performance.
The split between them signals how much of the outcome is attributed to the individual.
High variable proportions suit roles where the individual clearly drives the outcome — new business hunting with short cycles.
Lower variable proportions suit roles where outcomes depend on many factors — complex enterprise sales, account management, technical selling.
Mismatching these produces either unfair volatility or insufficient motivation.
What is measured
The decision that determines behaviour.
Revenue drives volume and encourages discounting to close.
Gross margin discourages discounting and requires the seller to have margin visibility, which many organisations do not provide.
New logos drives acquisition and neglects expansion within existing accounts.
Renewal and expansion metrics drive retention and can discourage difficult new business.
Which means plans measuring one thing produce that thing at the expense of others, reliably.
Accelerators
Higher commission rates above target.
Which motivates continued effort after target is reached, and it concentrates deals near period boundaries as sellers push to cross thresholds.
The resulting revenue pattern — heavy at quarter end — is a plan artefact rather than a customer behaviour, and it makes forecasting harder and gives buyers negotiating leverage.
Caps
Limiting total commission, generally introduced after someone earns an unexpectedly large amount.
Which is understandable and is consistently identified as damaging, since it removes the incentive precisely for the largest opportunities.
The better response to an outlier is generally to accept it and fix the plan for the following period, since capping mid-period destroys trust.
Quotas
Setting them well is genuinely hard.
Too high and the plan demotivates, since an unreachable target is equivalent to no incentive.
Too low and it costs money without driving effort.
The common guidance is that a substantial majority of the team should reach target, which is far more generous than many organisations set, and the reasoning is that a plan most people miss is a plan most people ignore.
Territory and allocation
Quota fairness depends on territory quality, which varies.
Which means identical quotas on unequal territories produce unequal outcomes unrelated to performance, and it is the most common source of perceived unfairness.
Assessing territory potential explicitly, and setting quotas proportionally, addresses it and requires data most organisations could produce and do not.
Clawbacks
Recovering commission when a customer cancels or fails to pay.
Which aligns sellers with customer quality and creates uncertainty about earned income.
Defined clearly and limited in duration, they are reasonable. Open-ended clawbacks are corrosive.
Changing plans
Frequent changes destroy the plan's ability to direct behaviour, since sellers stop believing it will persist.
Which means changes should be infrequent, announced with notice, and explained in terms of what the business is trying to achieve.
Mid-year changes that reduce earning potential are the most damaging action available to a sales leadership team, and they are taken regularly.
Team and individual
Purely individual incentives discourage collaboration, which matters where deals require several people.
Purely team incentives dilute individual accountability and allow free riding.
Most plans combine them, and the split should reflect how much the outcome actually depends on collaboration in that business.
Non-sales roles
Sales engineers, customer success and support all affect revenue outcomes.
Which raises the question of whether they should carry variable compensation, and practice varies considerably.
The general finding is that variable pay works where the individual has genuine influence over the outcome and produces resentment where they do not.
Communication
Plans that people cannot calculate for themselves do not motivate, since the connection between action and reward is unclear.
Which argues for simplicity over precision, and most plans err toward complexity because each addition addresses a specific concern.
A plan requiring a spreadsheet to understand is a plan that will be ignored in favour of whatever feels rewarded.
Ramp
New sellers cannot produce immediately, which means quotas should reflect a ramp period.
Which is standard and frequently set too short, producing early failures that were structural rather than individual.
Measuring how long existing sellers actually took to reach full productivity provides the figure.
Transparency
Plans differing between individuals without a stated basis generate resentment when discovered, and they are discovered.
Which argues for a consistent structure with differences based on stated criteria such as role and seniority.
Non-financial recognition
Public recognition, progression and interesting accounts all motivate alongside money, and organisations relying entirely on commission underuse them.
Which matters most for retaining strong performers, since the highest earners are generally motivated by more than the next percentage point.
Reviewing outcomes
Analysing what the plan actually produced — deal sizes, discount levels, timing, product mix — shows whether it drove the intended behaviour.
Which is the feedback loop that improves plan design and is rarely closed.