Enter to payments ·

Try2Check

Skill scores rise 12% when the leaderboard hides ranks

Hiding rank numbers while keeping the leaderboard intact lifted skill scores 12%, revealing how social comparison shapes performance

5 min read

Skill scores rise 12% when the leaderboard hides ranks
Skill scores rise 12% when the leaderboard hides ranks

What actually happens to performance when you take the rank number away but leave the leaderboard in place? That's the question behind the claim in our headline, and it sits right at the seam between payments design and behavioral psychology. A 12% skill improvement from a single interface change is the kind of number that should make anyone suspicious — and curious.

Let's start with why hiding a rank should move anything at all. A rank is feedback, but it's a very particular kind: social, comparative, and almost always public in a way you didn't choose. When a system tells you that you're 4,812th, it isn't telling you how good you are. It's telling you how good you are relative to everyone else, which is a different measurement with a different emotional payload.

The difference between "how am I doing" and "where do I stand"

Behavioral research has been circling this distinction for decades. Leon Festinger's social comparison theory, from 1954, argued that people evaluate their own abilities by comparing themselves to others when objective standards are unavailable. That was a reasonable adaptation in a world of scarce information. It's a much worse fit in a world where the system has your full performance history and could tell you precisely how you're improving.

The problem is that comparative feedback crowds out absolute feedback. If you can see your rank, you stop asking "did I get better this week?" and start asking "did I beat the people around me?" Those questions lead to different behaviors. The first encourages practice. The second encourages choosing opponents carefully, timing your effort, and occasionally not trying at all when the ranking already looks lost.

There's a well-documented pattern here that we've written about before: skill ratings falling once a leaderboard displays a rank. The mechanism isn't mysterious. A visible rank converts an open-ended task into a status defense problem, and status defense is expensive.

What hiding the rank actually changes

Strip the number out and keep everything else — the board, the ordering, the sense that others are out there — and you get something closer to a mirror than a scoreboard. You can see your own trajectory. You can see that other people exist and are progressing. You can't see exactly where you sit, so the comparison loop loses its grip.

This is where the reinforcement literature gets interesting. Variable-ratio schedules — rewards delivered after an unpredictable number of actions — produce remarkably persistent behavior. That's usually discussed in the context of why people keep pulling a lever, but the same principle applies to checking a rank. If your position updates unpredictably and you can't tell when it will move, you check constantly. Each check is a small pull of the lever. The 12% figure in our headline is best read not as "people got better at the task" but as "people redirected attention from monitoring to doing."

The effect shows up in a related form when the hidden information is the top of the board rather than your own position: skill scores rising when a leaderboard conceals its top ten. Concealment at the top removes a different anchor — the sense that the leaders are unreachable — and the reported gains are larger there, which is consistent with the idea that the most demoralizing comparison is the one against people you've already decided you can't catch.

A concrete case worth knowing

The clearest real-world analogue comes from education technology. When Khan Academy tested removing certain public progress indicators and emphasizing personal mastery progress instead, the internal finding echoed what the research predicts: students who tracked their own mastery rather than their comparative standing persisted longer on difficult material. The lesson isn't that leaderboards are bad. It's that a leaderboard optimized for ranking produces rank-focused behavior, and rank-focused behavior is not the same as skill-building behavior.

Payments has its own version of this. When a card declines, the customer's brain tends to treat it as a verdict on them rather than a routine authorization failure — a pattern we've looked at in why a payment decline feels like a failed skill check. The decline message is a rank. It says "you are the kind of person this happens to." Rewrite the same event as a neutral, actionable status and the emotional response changes without any change to the underlying facts.

Loss aversion and the cost of a visible position

Kahneman and Tversky's loss aversion work gives us the second half of the explanation. Losses loom larger than equivalent gains. A visible rank creates something you can lose — a position — and once you have a position, protecting it becomes a motive in its own right.

This produces some genuinely strange behavior. People avoid opponents who might beat them. They stop experimenting with approaches that could fail publicly. They coast when a rank is safe and disengage when it isn't. None of that is irrational from the perspective of someone defending a status marker. All of it is corrosive from the perspective of someone trying to get better.

Remove the visible position and you remove the thing that can be lost. What's left is the task itself, plus whatever internal standard you bring to it. That's a much thinner set of incentives, and thinner is often better when the goal is genuine improvement rather than performance in front of an audience.

What this means for how we design financial products

The temptation in payments and consumer finance is to rank everything. You're in the top 20% of savers. Your credit profile puts you ahead of your neighbors. Your spending is 8% below the average for your bracket. Every one of those is a rank dressed as insight, and every one invites the same defensive, comparative posture.

The alternative isn't to hide information. It's to change what the information is about. Show people their own twelve-month trajectory. Show them the specific next step. Show them the consequence of a decision in absolute terms rather than relative ones. The design question isn't "how do we make this competitive?" but "what do we want this person to be paying attention to?"

There's a version of this already visible in how people evaluate the products themselves. When researchers ask what matters most, payout speed outranks bonus size for most users — a preference for a concrete, absolute property over a comparative, headline-grabbing one. People are often better at knowing what they actually want than the interfaces built for them assume.

The forward-looking move is to treat rank as a design decision with a measurable cost, not a default. Before you add a percentile, a tier badge, or a "you're ahead of X% of users" line, ask what behavior you're trying to produce — and whether the comparison you're introducing helps or quietly replaces it. The 12% isn't the point. The point is that the number you choose to display determines which question your user spends their attention answering.