Test pay and recognition systems like medicines: small trials with pre-set measures

The problem: organisations introduce bonus schemes, targets and recognition programmes on confidence, and rarely test whether they work, or whether they have side effects.

The proposal: treat a new incentive scheme like a treatment. Try it first on a subset of teams, compare with similar teams that do not receive it, and decide in advance what outcomes will count as success, including quality, collaboration and staff turnover as well as output. Dan Pink's talk "The puzzle of motivation" is a well-known argument that, for work needing creativity, bigger rewards do not always produce better results. It is one view to test against your own workplace, not a verdict.

Who acts: a human-resources team designs the trial; an independent analyst, or a statistician inside the organisation, checks it.

Costs and trade-offs: trials take time, and employees may feel unfairly treated if some groups get a benefit first. Transparency about the purpose, and giving everyone the scheme if it works, address this.

First step: pick one scheme and one measure and run a six-month comparison.