Give your best engineer a pay rise and a fancier title, and you may keep them from quitting, without making them one bit more motivated on Tuesday morning. That gap, between not-unhappy and genuinely driven, is the whole subject of motivation theory, and the models that explain it best are not always the ones on the slide deck.

The quick version

  • Maslow's hierarchy (needs stacked from food to self-actualisation) is intuitive and useful as a vocabulary, but the strict "climb one level at a time" version was never well supported by data.
  • Herzberg split work into hygiene factors (pay, conditions, policy, fixing them stops dissatisfaction but doesn't motivate) and motivators (achievement, the work itself, growth). The split is a useful lens; the underlying study has real method flaws.
  • Self-Determination Theory is the best-evidenced of the three: people are driven when three needs are met, autonomy, competence and relatedness, and tangible rewards can quietly crowd out the intrinsic interest you were relying on.
  • The move is the same across all three: don't try to buy motivation. Remove the irritants, then make the work itself meaningful, masterable and self-directed.

The idea in depth: Maslow and Herzberg, famous, useful, and shakier than you think

Most managers meet motivation through two mid-century models. Both are worth knowing. Both are weaker as science than their fame suggests, which is exactly why you should hold them as lenses rather than laws.

Abraham Maslow's hierarchy of needs first appeared in "A Theory of Human Motivation" (Psychological Review, 1943). His claim was that human needs arrange themselves in rough tiers, physiological, safety, belonging, esteem, and finally self-actualisation, and that lower needs broadly take priority until they're reasonably met. It is a genuinely useful way to notice that a person worried about losing their job is not in the headspace to chase a stretch goal. But the popular pyramid (Maslow himself never drew one) oversells a rigid ladder. The most thorough review, Mahmoud Wahba and Lawrence Bridwell's "Maslow Reconsidered" (Organizational Behavior and Human Performance, 1976), examined the available studies and found little support for the strict ranking or the "satisfy-then-ascend" mechanism, self-actualisation was about the only tier that behaved as predicted.

So the move is: use Maslow as a diagnostic prompt, not a sequence. Before you reach for an inspiring vision, check whether a more basic need is unmet, is the role insecure, is the person isolated on the team, are they underwater on workload? Fix the floor before you decorate the ceiling. Just don't assume people climb in a tidy order, because the evidence says they don't.

Frederick Herzberg's two-factor theory made a sharper, more practical claim. From interviews asking people to recall times they felt exceptionally good or bad about work, he concluded that the things that cause dissatisfaction are different in kind from the things that cause satisfaction. Pay, company policy, supervision and working conditions he called hygiene factors: get them wrong and people are miserable; get them right and you reach merely neutral. Real motivation came from motivators, achievement, recognition, responsibility, the work itself, and growth. His argument is laid out in "One More Time: How Do You Motivate Employees?", first published in Harvard Business Review in 1968 and reprinted so often it became one of HBR's all-time bestsellers.

"Forget praise. Forget punishment. Forget cash. You need to make their jobs more interesting.", the line Harvard Business Review ran atop Herzberg's article, and a fair distillation of where he landed.

So the move is: treat pay and conditions as table stakes, then redesign the job to add motivators. Herzberg called this job enrichment, giving people more ownership of a whole piece of work, more autonomy in how they do it, and direct feedback on the result, rather than just shuffling tasks sideways. The honest limitation: the theory leans heavily on Herzberg's particular interview method, and when researchers used other methods the clean two-factor split often blurred, the same factor (pay especially) can show up on both sides. So keep the practical insight (you can't pay your way to motivation) and hold the rigid two-bucket model loosely.

flowchart LR
  A(["Hygiene factors
pay · conditions · policy
· supervision"]) --> B(["Fix them and you reach
NEUTRAL, not motivated,
just not unhappy"])
  C(["Motivators
achievement · the work itself
· responsibility · growth"]) --> D(["Add them and you reach
genuine MOTIVATION"])
  B --> E(["The two are different in kind
you need both jobs done"])
  D --> E
Herzberg's core claim: removing dissatisfiers only gets you to neutral. Motivation comes from a different set of factors entirely. Leaders Loop

The idea in depth: Self-Determination Theory, the one with the evidence

If you only internalise one model, make it this one. Edward Deci and Richard Ryan's Self-Determination Theory (SDT) has been built and tested across five decades, and it answers the question the older models gesture at: what makes work intrinsically motivating? Their answer is three basic psychological needs, summarised in Ryan & Deci's overview in American Psychologist (2000):

Autonomy, a sense that your actions are your own, not imposed. Competence, the feeling of getting good at something that matters and seeing yourself grow. Relatedness, feeling connected to and valued by the people around you. Where these three are supported, SDT finds higher-quality motivation, better performance and better wellbeing; where they're thwarted, by controlling managers, pointless work or isolation, motivation curdles into compliance or apathy.

SDT's most counter-intuitive, and most replicated, finding concerns rewards. In Deci's original 1971 experiments, people paid to solve puzzles they'd previously enjoyed for their own sake became less likely to play with them once the money stopped, the reward had quietly reframed the activity from "something I do because it's interesting" to "something I do for cash." Deci, Richard Koestner and Ryan later pooled the evidence in a meta-analysis of 128 experiments, "A Meta-Analytic Review of Experiments Examining the Effects of Extrinsic Rewards on Intrinsic Motivation" (Psychological Bulletin, 1999). Across the studies, tangible rewards that were expected and tied to doing a task significantly undermined free-choice intrinsic motivation, an effect they call the undermining or overjustification effect. Verbal recognition, by contrast, tended to boost intrinsic motivation, because it feeds competence without feeling controlling.

This is the heart of the intrinsic vs extrinsic distinction. Extrinsic motivation comes from outside the task (pay, grades, threats, prizes); intrinsic motivation comes from the task itself. Extrinsic levers aren't evil, for dull, mechanical work they may be exactly right, but they're blunt, and used carelessly on work people already cared about, they can erode the very interest you wanted to amplify.

So the move is: engineer autonomy, competence and relatedness into ordinary work, and reach for rewards last. Give people genuine choice in how (and sometimes what) they do, that's autonomy. Set goals at the edge of their current ability with fast, honest feedback, that's competence. Make sure no one is rowing alone, that's relatedness. And when you do recognise someone, prefer specific, informational praise ("the way you handled that escalation kept the account") over a generic bonus, because the first builds intrinsic drive and the second can crowd it out. The honest limitation: most SDT lab work undermines an already intrinsically interesting task, and pay obviously still matters in the real world, nobody works for autonomy on an empty stomach. SDT tells you not to rely on rewards to manufacture motivation that should come from the work; it does not tell you to underpay people.

Pay keeps people from leaving. It doesn't make them care. Those are two different jobs, and they need two different tools.

flowchart TD
  A(["What kind of work
is it?"]) --> B(["Already interesting /
creative / judgement-heavy"])
  A --> C(["Routine / mechanical /
clearly measured"])
  B --> D(["Support autonomy,
competence, relatedness
avoid controlling rewards"])
  C --> E(["Extrinsic incentives
can work, keep them
fair and simple"])
  D --> F(["First fix the hygiene
floor (pay, conditions),
THEN add motivators"])
  E --> F
A practical synthesis: match the lever to the work, and treat fair pay as a floor you build motivation on top of, not the motivation itself. Leaders Loop

A worked example

Sofia runs a support team of eight. Tickets are being closed on time, but the energy has gone flat: her two strongest agents have stopped suggesting improvements, and one has started asking about "what's next" in a way that sounds like one foot out the door. HR's instinct is a retention bonus. (This scenario and any figures are illustrative.)

Read it through the three models. Through Herzberg, the bonus is a hygiene move, it might stop someone leaving, but it won't put the spark back; the flatness is a missing-motivator problem. Through Maslow, the basics are covered (the floor is fine), so the unmet need is higher up, esteem and growth. Through SDT, all three needs are quietly starved: the work has become rote (no competence stretch), it's tightly scripted (no autonomy), and the agents process tickets in parallel silence (thin relatedness). A cash bonus addresses none of that, and SDT's evidence warns it could even reframe good work as something done only for the money.

So Sofia redesigns the work, not the wallet. She gives each senior agent ownership of a recurring problem category, authority to fix root causes, not just close tickets (autonomy + competence). She starts a short weekly session where the team brings the gnarliest case and solves it together (relatedness + mastery). She replaces the generic "great job team" with specific, informational recognition tied to real outcomes. Pay stays competitive, that's the floor, and she checks it's genuinely fair, but it is no longer asked to do a job it was never able to do. Within a quarter the improvement ideas come back, because the work itself once again asks something of them.

Frequently asked questions

So does money not motivate people at all?

It motivates, but narrowly and with a ceiling. Fair, competitive pay is essential, underpay people and nothing else you do will land. But once pay is fair, more money is a weak and short-lived lever for the engaged, creative work most knowledge teams do, and Herzberg's framing explains why: pay mostly removes dissatisfaction rather than creating motivation. For routine, well-measured tasks, incentives work better. The error is using cash to fix a meaning problem.

If Maslow and Herzberg aren't well supported, why learn them?

Because they're useful vocabulary and they're everywhere, you'll hear them in every management conversation, so you need to know both what they offer and where they're thin. Maslow reminds you to check the floor before chasing the ceiling; Herzberg reminds you that fixing complaints isn't the same as creating drive. Treat them as lenses that sharpen attention, and lean on Self-Determination Theory when you want the model with the strongest evidence behind it.

What's the difference between intrinsic and extrinsic motivation, plainly?

Intrinsic motivation is doing something because the activity itself is rewarding, interesting, absorbing, meaningful. Extrinsic motivation is doing it for an outcome outside the task, pay, a deadline, avoiding trouble. Both are real and both have a place. The catch SDT surfaces is that piling extrinsic rewards onto already-intrinsic work can crowd the intrinsic motivation out, so use external levers deliberately, not by default.

Can I really make a boring job motivating?

Partly. You can't always change the tasks, but you can change the conditions around them: give people more say in how the work is done, make progress visible so they feel competence, and connect them to the people the work helps so it carries meaning. Herzberg's job enrichment and SDT's three needs point the same way, add ownership, mastery and connection. Where the work is genuinely mechanical, be honest and pay fairly for it rather than pretending it's a calling.

Where does this break down?

In three places worth naming. First, individual and cultural variation, what reads as motivating autonomy to one person can feel like being abandoned to another. Second, the lab-to-life gap, much of the reward-undermining evidence comes from short experiments on already-enjoyable tasks. Third, the floor problem, none of this substitutes for fair pay, security and decent conditions. Use these models to design better, then watch what actually happens with your people and adjust.

Related in the Toolkit

Where to go next