Sales contests that grouped employees with similar performers produced better average results than a single contest for an entire sales force in a study published in the Journal of Marketing Research.
The researchers compared a contest open to all salespeople with contests divided into smaller segments, as well as a control group without a contest. In the main field experiment, the single contest did not outperform the no-contest condition on average.
The research suggests that the chance of winning can affect whether an incentive motivates a salesperson. It does not establish that every team will benefit from a contest or that sales targets can substitute for good management.
One leaderboard or several?
Raghu Bommaraju, S. Arunachalam and Sebastian Hohenberg studied two ways of organizing sales competitions. In the first, all salespeople competed against one another. In the second, the company divided salespeople into comparable segments and they competed within their own groups.
The main field experiment included two contest groups and a group without a contest. The authors supplemented it with two further field studies and two experimental studies.
In the research settings, the segmented contests improved average performance and performance across segments. The researchers also compared segmented contests with one-on-one competitions between salespeople in the same segment and found stronger results for the segmented format.
The paper appeared online in 2024 and was published in the journal’s June 2025 issue. Its findings concern the designs and sales forces examined by the authors.
Why the perceived chance of winning matters
A leaderboard can be motivating when employees believe their effort might change the outcome. If a small group of leading sellers appears certain to win, others may decide that the prize is out of reach.
The authors found evidence that a higher perceived chance of winning and greater motivation helped explain the stronger performance in segmented contests.
Grouping employees does not remove every source of unfairness. Salespeople may have different territories, customer accounts, products or levels of experience. Those differences can affect sales even when people have similar past results.
A company choosing groups would need to decide which measures make comparisons reasonable and how often to review them. The study does not provide a universal formula for every sales organization.
Designing an incentive that fits the work
Before launching a contest, managers can define the behavior they want to encourage. Total sales may be an appropriate measure in some settings, while others may need to consider the quality of accounts, customer retention or the margin on a sale.
Focusing on one metric can create unwanted incentives. A contest based only on revenue, for example, could reward large discounts that bring in orders but leave little profit.
A profit margin measures the share of revenue remaining after specified costs. Managers comparing sales outcomes should be clear about which costs their measure includes.
Businesses can also set transparent eligibility rules, explain how results are calculated and check whether employees have a realistic opportunity to participate. A prize’s cost belongs in the calculation alongside any extra revenue.
Test outcomes beyond the leaderboard
A company can compare a contest period with a suitable control group or, where feasible, randomly assign comparable teams to different approaches. The comparison should account for seasonal demand, territory changes and other events that affect sales.
It can also track returns, customer complaints and repeat purchases. A short-term increase in orders is less attractive if customers cancel later or employees feel pressured to sell unsuitable products.
The research adds evidence that the structure of a competition can matter as much as the prize. For a manager, the decision starts with whether employees can influence the result and whether the rewarded sales are valuable to customers and the business.