When Can a Reward Make Performance Worse

Start talking

Start talking: first response

A support team earns a bonus for closing more requests. Closures increase, but customers complain that their problems remain. Did the incentive work?

Take it further

What would make your answer different for the manager, the staff and the customer?

Clip A

Pink uses the candle problem to challenge a simple prediction about performance rewards. Compare the two versions of the task and the results he reports; avoid treating one striking result as a rule for every kind of work.

Watch for: What change in the task reverses the reported effect of the reward?

Academic Vocabulary

performance incentive

A reward or consequence linked to an assessed result or behaviour.

offer a performance incentive

The firm offered a performance incentive for completing more requests.

functional fixedness

Difficulty recognising an object’s possible uses beyond its familiar function.

overcome functional fixedness

The team overcame functional fixedness by treating the container as a support.

intrinsic motivation

Motivation arising from interest, satisfaction or meaning in an activity itself.

support intrinsic motivation

Giving staff meaningful problems may support intrinsic motivation, although working conditions still matter.

insight

A new way of understanding a problem that reveals a possible solution.

reach an insight

The useful insight was that the apparent packaging could become part of the solution.

Original source

The puzzle of motivation , Dan Pink · TED · 01:31–07:04 · Source and credit

Clip A quiz: 6 questions

Choose one answer for each question, then read the feedback.

1. What is the candle problem’s goal?

Make the candle burn as quickly as possible.Identify which brand of candle produces the least wax.Attach the candle to the wall without wax dripping onto the table.

2. What must participants recognise to overcome functional fixedness here?

The box can serve as a platform, not only a container.The candle must keep every object in its original position.The matches provide a permanent adhesive by themselves.

3. How did the rewarded group perform in the first experiment Pink describes?

It finished faster in every individual case.It was not timed, so its performance could not be compared.It took longer on average.

4. What changed in the second version?

The task no longer involved a candle or a wall.The drawing pins were presented outside the box.Participants were given the answer before starting.

5. What result does Pink report for the easier-to-see version?

Rewards again produced the same slower result.The rewarded group performed faster.Both groups were deliberately prevented from finishing.

6. How does Pink explain the contrasting reward effects?

Narrowed attention can help clear tasks but hinder a needed change of perspective.Rewards always reduce effort regardless of the task.Insight requires the complete absence of any goal.

Grammar in a minute: A what-cleft for diagnosis

What + clause + be highlights the element that follows. It is useful for correcting a vague diagnosis of a problem.

You say: Explain what a difficult work task needs beyond greater speed.

One possible response

What the task requires is a careful diagnosis, not simply a faster reply.

Useful English

Useful English: speak and compare

Say your response before opening a model.

The result depends on the kind of task.

Limit a general claim about performance.

You say: Respond to a colleague who says a bonus must improve every job.

One possible response

The result depends on the kind of task. A clear repeated procedure may respond differently from a problem that needs a new interpretation.

That comparison changes one crucial feature: …

Identify what makes two cases informative.

You say: Explain why the two candle-task versions are worth comparing.

One possible response

That comparison changes one crucial feature: how easily the box’s alternative use can be recognised. The contrast helps us question a universal prediction about rewards.

Critical Thinking: 3 questions

Choose one question, then follow up, challenge and revise. Use the others for another angle.

Does the candle experiment justify abolishing all workplace bonuses?
Help me start
Distinguish the reported task effect from a general employment policy.
One possible response
No. It gives a reason to question an automatic prediction, especially where reframing matters. A workplace decision also needs evidence about the actual task, the reward design and other consequences. The second version already warns against treating every task alike.
Take it further: What kind of job would be a poor match for this experiment?
C2 challenge
State a cautious transferable hypothesis and a result that would count against it.
How would you distinguish increased effort from improved problem-solving?
Help me start
Consider time spent, attempted approaches and solution quality.
One possible response
A person may work harder while pursuing an unproductive approach. I would examine successful outcomes and the range of strategies used, not only time or visible intensity. The relevant measure depends on whether the task rewards persistence, insight or both.
Take it further: Could taking longer sometimes signal better performance?
C2 challenge
Propose two measures that could move in opposite directions and explain why.
Is pressure necessarily harmful for creative work?
Help me start
Separate a manageable deadline from a narrowly rewarded target.
One possible response
Not necessarily. Some structure can focus attention and support completion. I would distinguish productive boundaries from pressure that makes people avoid exploration or conceal failure, then ask which conditions help this team produce useful work.
Take it further: What sign would tell you the pressure has become counterproductive?
C2 challenge
Defend a deadline while identifying a part of the task that needs protected exploration.

Clip B

Grant and Gneezy examine how rewards influence both behaviour and the meaning of good work. Their discussion moves from unintended effects to practical design choices, including collaboration, innovation and a bus-driver example.

Watch for: What message can a reward send when it conflicts with a manager’s stated priorities?

Academic Vocabulary

metric

A defined measure used to assess an activity or result.

choose a performance metric

The chosen performance metric should reflect the outcome the service needs.

misaligned goal

An objective or reward target that does not support the wider intended purpose.

identify a misaligned goal

Counting closures created a misaligned goal when customers needed lasting solutions.

signal

Information conveyed by an action that influences what people believe or expect.

send a conflicting signal

Praising collaboration while rewarding only individual totals sends a conflicting signal.

pilot study

A limited initial study used to learn how an intervention works and inform further decisions.

run a pilot study

The team will run a pilot study before extending the new bonus arrangement.

Original source

How do incentives really work? with Uri Gneezy | Re:Thinking with Adam Grant · TED Audio Collective · 04:39–10:01 · Source and credit

Clip B quiz: 6 questions

Choose one answer for each question, then read the feedback.

1. Why does Gneezy reject the claim that one failed scheme proves incentives do not work?

A poorly designed example does not settle the whole category.Any reward always produces the intended outcome.Failure is irrelevant when the reward is expensive.

2. What does he recommend when intuition about an incentive is uncertain?

Keep the first design unchanged to protect managerial confidence.Assume a larger reward must fix the problem.Test its effects and modify the design if needed.

3. What contradiction does Grant identify in the collaboration example?

Managers provide a shared goal and recognise shared contributions.Employees are given no information about any desired outcome.Managers request teamwork while rewarding only individual results.

4. Why can punishing every failed innovation undermine the stated aim?

Trying unfamiliar ideas carries a possibility of failure.Innovation is possible only when outcomes are guaranteed in advance.Any failed attempt demonstrates that the employee ignored the task.

5. What does paying a bus driver per passenger communicate in Gneezy’s example?

Passenger volume is what the organisation actually rewards.Safety automatically becomes the only relevant target.Customer satisfaction has been measured directly by the payment.

6. Why does removing an explicit bonus not create a situation without incentives?

The absence of a bonus guarantees that every goal is aligned.The existing setting still contains motives and consequences.People lose all reasons to work unless a bonus is offered.

Grammar in a minute: An it-cleft for the real signal

It is + focused element + that highlights the source of an effect. Keep the following clause complete.

You say: Emphasise the part of a scheme that conflicts with its stated aim.

One possible response

It is the individual ranking that discourages staff from spending time helping colleagues.

Useful English

Useful English: speak and compare

Say your response before opening a model.

What the scheme rewards is …

Name the behaviour the system actually selects.

You say: Diagnose a bonus based only on completed support requests.

One possible response

What the scheme rewards is recorded completion. If lasting resolution is the goal, we need to check whether the count captures it.

Before scaling this, I would want to see …

Connect a wider rollout with relevant evidence.

You say: Propose evidence for extending a redesigned team reward.

One possible response

Before scaling this, I would want to see better resolution quality without a rise in hidden delays. A higher completion count on its own would not be enough.

Critical Thinking: 3 questions

Choose one question, then follow up, challenge and revise. Use the others for another angle.

Should managers reward a sensible attempt that fails?
Help me start
Distinguish the decision process, learning and the final outcome.
One possible response
Sometimes. If the organisation needs exploration, a well-chosen attempt can produce useful learning despite failure. I would distinguish that from careless execution and make the evaluation criteria clear enough that recognising learning does not become rewarding every unsuccessful action.
Take it further: What evidence would distinguish those two cases?
C2 challenge
Defend accountability without creating an incentive to hide uncertainty.
Would adding a quality score automatically fix a quantity bonus?
Help me start
Consider measurement, controllability and possible gaming.
One possible response
It could improve alignment, but the quality score might be noisy or easy to manipulate. I would ask whether employees can influence it fairly and whether it captures the customer’s actual outcome. Adding a second number does not remove the design problem.
Take it further: How might people improve the score without improving the service?
C2 challenge
Propose one safeguard and explain the new burden it creates.
Can a team bonus damage collaboration even when collaboration is the aim?
Help me start
Think about unequal effort, perceived fairness and information sharing.
One possible response
It might, if people believe contributions are hidden or free-riding is rewarded. A team measure can encourage shared responsibility, but the work and feedback arrangements also matter. I would examine how colleagues recognise and support each other’s contributions.
Take it further: What would help a high contributor accept the scheme?
C2 challenge
Compare individual and collective accountability without promising a perfect measure.

Mastery

Mastery: 6 questions

Original cases. Choose, then explain one answer aloud.

1. Fictional team closes more requests after a bonus, but many are reopened. What is the clearest diagnosis?

The bonus had no effect on behaviour.Every reopened request proves deliberate dishonesty.The rewarded count may diverge from lasting resolution.

2. A task becomes routine after a clear procedure is introduced. What does Pink’s comparison suggest investigating?

Whether the same incentive now has a different effect.Whether rewards must remain harmful because they were harmful once.Whether task structure is irrelevant to performance.

3. Managers ask for long-term research but promote only on quarterly output. What should they examine?

Only whether employees can repeat the mission statement.Only whether the reward amount is sufficiently large.The conflict between their stated time horizon and the reward horizon.

4. A trial improves one team’s output during an unusually quiet month. Which comparison would make its result more informative?

Comparable work and conditions without the changed scheme.The team’s confidence that the new rule was helpful.Only a second presentation of the same output total.

5. A quality score penalises staff for product defects they cannot control. What issue does this create?

Staff should stop reporting defects to protect the score.The reward may conflate employee performance with external causes.The score must be fair because it concerns quality.

6. A firm removes bonuses but retains public rankings and promotion targets. What follows from Gneezy’s argument?

The firm has eliminated every external motivator.All remaining motivation is necessarily intrinsic.Significant incentives remain in the work environment.

Grammar in a minute: A paired comparative

The more …, the more … links changes in two quantities or tendencies. Make the relationship meaningful rather than assume it is universally true.

You say: Describe a possible unintended relationship in a reward scheme.

One possible response

The more we emphasise individual totals, the less time people may devote to helping colleagues.

Repair the support-team bonus

Integrated spoken task: decide and revise

Fictional ClearDesk pays a bonus for the number of requests each adviser closes. Recorded output has risen, but complaints and repeat contacts have increased.

Agree a replacement scheme and a small trial. State the desired customer outcome, one way the scheme could be gamed and the evidence that would change your decision.

Open only your own role. Share its information by speaking.

Operations manager

Reduce the waiting list while protecting the service’s quality.

The board expects an immediate plan. A two-team trial is possible, but simply abandoning all targets would leave the backlog unmanaged.

Experienced adviser

Make the scheme reflect the work and support useful cooperation.

Helping a colleague reduces an adviser’s own closure count. Complex problems also take much longer than routine ones, although every closure earns the same credit.

Give an initial recommendation and its strongest reason.
Exchange private information; challenge one assumption and answer the objection.
Hear the new information, then revise or defend the decision with a stated reason.

New information: revise your decision

A new self-service feature removes most routine requests, leaving fewer cases that require more diagnosis.

Revise the target and trial for the changed task mix. Which old comparison is no longer fair?

A strong response

Use a relevant idea from each clip.
Answer the other person’s strongest objection with a reason.
Explain precisely how the new information changes, or does not change, your decision.

I would pause the raw individual closure bonus and trial a team measure of lasting resolution, with a separate view of waiting time. The self-service change means the remaining cases are harder, so comparing their speed with the old mixed workload would be misleading. Pink’s contrast suggests that task demands can change an incentive’s effect. Gneezy’s account also shows why the payment must support the outcome we actually value. We should recognise useful collaboration and track repeat contacts without blaming advisers for unrelated product defects. I would keep the scheme only if quality improves without unacceptable delay.

Spoken exit

Your spoken exit

Name the behaviour a reward measures, the outcome the organisation wants and one reason they may differ.

One possible response

A closure bonus measures recorded completions, while the organisation wants solved problems. Staff may close requests quickly without producing a lasting resolution.

Your takeaway

One-minute takeaway

Choose one answer. Then explain your choice in one sentence aloud.

When Can a Reward Make Performance Worse?

Start talking

Clip A

Useful English

Clip B

Useful English

Mastery

Repair the support-team bonus

Spoken exit

Your takeaway

Lesson aims

Practice audio