Why your reward system needs the wrong kind of skill to feel fair

You know that hollow feeling when you crush a work task but it just doesn't hit the same as a lucky guess? Or when you spend hours perfecting a website feature, only to feel strangely flat after launch? It’s a specific kind of cognitive dissonance. We’ve built reward systems — in business, in software, in our own habits — that are designed to feel fair, but they often fail because fairness, it turns out, requires a little bit of the wrong kind of skill. The skill of embracing uncertainty, not just eliminating it.

The fairness paradox: effort vs outcome

Let’s start with a simple truth from behavioural economics. Humans have a deep, almost obsessive need for effort to correlate with outcome. We call this “procedural fairness.” If I work hard, I should get a proportional reward. If I slack off, I shouldn’t. This is the bedrock of most workplace incentives, project management, and yes, website development.

But here’s the rub. The real world, especially the digital world, is deeply stochastic. A beautifully coded landing page can fail because of a random algorithm update. A scrappy, half-baked competitor can go viral through sheer dumb luck. When that happens, our brain screams “unfair.” We’ve been trained by the reward system — the dopamine loop of completing a ticket, hitting a KPI, shipping a feature — to expect a linear payout.

Daniel Kahneman’s work on “System 1” and “System 2” thinking is relevant here. System 1 is fast, intuitive, and emotional. It loves the immediate hit of a clear win. System 2 is slow, deliberate, and logical. It understands long-term strategy. A reward system that only feeds System 1 — the clean, predictable win — feels fair only as long as the world cooperates. The moment randomness enters (a client changes their mind, a competitor launches a cheaper product), the system breaks. It wasn't built for the mess.

Variable-ratio reinforcement: the dirty secret of engagement

This is where we need to borrow a concept from a completely different field: operant conditioning. B.F. Skinner’s work on variable-ratio reinforcement is often misunderstood as being about “addiction.” In reality, it’s about sustainable engagement.

A fixed-ratio reward (get a cookie every time you finish a chapter) is predictable. It feels fair in a transactional sense, but it quickly becomes boring. You hit a plateau. The brain stops caring. A variable-ratio reward (get a cookie sometimes, but you never know when) is infinitely more engaging. The uncertainty itself becomes part of the reward.

Here’s the key insight for your business: A reward system that feels fair must include a controlled dose of uncertainty. Not the uncertainty of failure — that’s just anxiety. The uncertainty of when the reward comes, and how big it will be.

Consider a common web development scenario: A/B testing. You run a test on a call-to-action button. You expect a clear winner. Instead, the results are a statistical tie. The “fair” response is to call it inconclusive and move on. The smart response is to realise that the uncertainty is data. The reward isn’t the winning variant; the reward is the learning from the tie. Most teams hate this. They want the clean win. But the teams that build reward systems that celebrate the messy, uncertain learning process (not just the clean outcome) are the ones that build better products.

The concrete example: the “near miss” in product design

Let me ground this in a specific study. Natália Kocsel and colleagues (2019) looked at the neural response to “near misses” in a non-gambling context. They found that a near miss — almost hitting a target, almost achieving a goal — activates the same reward circuitry in the brain as a full win, but with a crucial difference: it also activates regions associated with learning and motivation.

In a business context, we treat near misses as failures. “We almost closed that client.” “We almost hit the conversion target.” “The feature was almost perfect.” Our reward system punishes the near miss because it fell short of the fixed-ratio goal.

But what if you redesigned your reward system to specifically reward near misses? Imagine a development team that celebrates a feature that got 90% of the way there but uncovered a critical usability flaw. The “failure” is actually a high-value learning event. The reward isn’t the launch; it’s the discovery. This feels deeply unfair to the part of our brain that wants the clean win. But it’s far more effective for long-term growth.

This is the wrong kind of skill I’m talking about. The skill of sitting with the near miss. The skill of extracting value from the almost-win. The skill of not needing the dopamine hit of the finished checkbox.

Building a reward system that trusts the mess

So how do you actually build this into your workflow? It’s not about removing goals. It’s about layering in a second, less obvious reward track.

Track 1: The Predictable Win (Fixed-Ratio) This is your baseline. Ship a feature, get a thank you. Hit a deadline, get a break. Close a deal, get a commission. This is the “fair” part. It provides stability. It prevents burnout. It’s the floor.

Track 2: The Discovery Reward (Variable-Ratio) This is where the magic happens. This track rewards unexpected outcomes. It rewards the hypothesis that was wrong but taught you something. It rewards the client meeting that went sideways but uncovered a new need. It rewards the code that broke in a surprising way.

How do you operationalise this? Simple. Create a “Stochastic Kudos” channel in your team’s communication tool. It’s not for wins. It’s for surprises. Someone finds a bug that reveals a deeper pattern? Kudos. A test result is confusing? Kudos. A client gives feedback you didn’t expect? Kudos. The reward is social recognition for embracing the variable.

This feels unfair to the traditionalist. “Why are we celebrating confusion?” Because confusion is the raw material of insight. A reward system that only celebrates clarity is a system that breeds brittle, fragile work.

The forward-looking close

The next time you feel that hollow satisfaction after a “clean” win, ask yourself: was that win real, or was it just predictable? The most resilient systems, whether they’re websites or teams, are built on a foundation that can handle the random. They don’t just tolerate uncertainty; they build reward loops around it.

Start small. Pick one project this week. Don’t ask “Did we win?” Ask “What surprised us?” Then reward the surprise, not the win. It will feel wrong. It will feel unfair. That’s the point. That’s the skill you need to build. Because in a world that is fundamentally uncertain, the only truly fair reward system is one that pays out for the ride, not just the destination.