Rewards and Recognition That Don't Backfire
The most instructive recognition program I ever watched was one that backfired. A monthly award for whoever resolved the most incidents, well-intentioned, promptly created an incentive to be the firefighter and none to be the person whose careful work prevents fires. Prevention is invisible; heroics are legible. Within two quarters, the quietly reliable engineers, the ones whose modules never woke anyone at 2 a.m., understood exactly where they stood. What you reward is what you get more of, and the program was rewarding smoke.
Recognition is a system with failure modes, and I have collected several first-hand. The design rules below are what remains after the mistakes.
Reward what you want repeated, and check what that actually is
Before praising anything, ask what behavior the praise multiplies. Celebrating the weekend warrior who saved the release multiplies weekend heroics, and quietly signals that planning well enough to not need them is worth less. My counterweight is deliberate: publicly credit the undramatic wins, the migration that went so smoothly nobody noticed, the module that has not paged anyone in a year, the estimate that was honest and therefore unpopular. Boring excellence needs a publicist, because it has no natural one.
Recognize by contribution, not by visibility
Left uncorrected, recognition flows toward proximity: whoever demos, whoever speaks in the steering meeting, whoever the sponsor happens to see. The engineer deep in the data layer, the tester whose scenarios caught what would have been the worst incident of the year, are structurally invisible. Part of the PM's actual job, as the person with the widest view of who did what, is being the accuracy layer, routing credit to where the work happened, by name, in the rooms where it counts. This is also, not incidentally, one of the few reward powers a matrix PM genuinely holds.
Specific beats generic by an order of magnitude
"Great job everyone" costs nothing and is worth what it costs. Specific recognition, "the rollback plan you insisted on is the reason today was an inconvenience instead of a crisis", does three things at once: it proves someone actually saw the work, it teaches the room what good looks like, and it survives in memory. My rule for praise is the same as for quality standards: if it could apply to anyone, it lands on no one.
Match the currency to the person
The motivation theories post covers why the same reward reads differently per person, and recognition is where that cashes out. Public praise delights some people and genuinely mortifies others; some value the growth opportunity far above the bonus; for some the most meaningful reward available is trust in the form of a harder problem. Getting this wrong with good intentions, the public spotlight on someone who hates spotlights, still counts as getting it wrong. Knowing your people's currencies is not extra credit; it is the prerequisite.
Fairness is the load-bearing wall
One visibly unfair award damages more than ten good ones build, because recognition systems are read by everyone who didn't receive them. The moment credit appears politically routed, to favorites, to the loudest, to whoever the manager lunches with, the entire mechanism converts from motivator to cynicism generator. Herzberg applies here too: fair recognition motivates, but unfair recognition actively poisons, and the poison spreads to people who were previously content.
The cheap, durable version
Everything above works without budget. A recognition system in its minimal viable form is one habit: regularly, specifically, fairly, and in the right currency, make sure good work is seen by someone who matters, including, sometimes, only the person themselves in a one-on-one. Money and awards can amplify that habit. They cannot replace it, and every attempt I have seen to substitute budget for attention produced exactly what it invested: expensive indifference.
Nguyễn Hải Nam
Project Management Lead. 16+ years from code to delivery. PMP®. Writing here about project management and engineering.