Gamification that changes behavior

notes.

Gamification that changes behavior

Source: I Studied 500+ Gamified Apps (Here’s What Actually Works), Tim Gabe, 12:33, uploaded 2026-05-11.

Good gamification gives a useful behavior a visible shape. It makes progress, competence, and social context easier to perceive. The video’s seven patterns are less a menu of tricks than a sequence of tests for whether the game layer still serves the underlying activity.

The scoreboard without a game

Points, badges, and leaderboards are the first things many teams add and some of the best documented failures. LinkedIn retired its Community Top Voice gold badges after badge-seeking increased the quantity of posts rather than the quality of expertise. Foursquare removed mayorships and badges after they encouraged check-ins without improving the discovery behavior the business needed. Google News abandoned a badge system for the same general reason.

Yu-kai Chou’s formulation carries the argument: points, badges, and leaderboards are the scoreboard, not the game. A scoreboard can expose the state of baseball; it cannot make an empty stadium feel like baseball.

Strava and engineered winnability

Strava appears to contradict the critique because its segments are leaderboards. The difference is scale. A user-defined hill on a morning route creates a local competition sorted into plausible cohorts rather than one global ranking that almost everyone must lose. The video cites winnability as the strongest predictor of competitive motivation.

The app also maintains an unusual relation between interface time and real activity: the video reports about one hour of exercise for every two minutes spent in the app. Kudos attach social acknowledgement to an actual run or ride, and are said to increase future running frequency. The platform is not merely counting engagement with itself.

The S-curve and Habitica

This might suggest adding more good mechanics, but the third pattern is a limit. A cited 2025 study describes feature richness as an S-curve: engagement improves as useful mechanics accumulate, reaches a peak, then falls as the system becomes work.

Habitica provides the extreme case. Tasks become quests, habits become character statistics, and missed work damages health points. In the study described by the video, every participant experienced some counterproductive effect. People became absorbed in maintaining the game representation of productivity while the actual work receded. Quests, currencies, stats, badges, challenges, and leaderboards can become cognitive overload wearing an engagement costume.

Streaks become obligations

The emotional engine of a streak changes as it grows. “I want to do this” becomes “I cannot miss today.” The video links Snapchat streak frequency among Belgian adolescents with fear of missing out, problematic smartphone use, and reduced self-control, then notes litigation in Nevada and forthcoming European scrutiny of addictive mechanics.

This does not make every streak abusive. Duolingo lets people choose a goal and obtain streak freezes. The pressure comes with agency, repair, and an escape hatch. A streak that cannot be paused, influenced, or left is the more dangerous design.

Anticipation instead of threatened loss

Variable reward magnitude produces a different engine. In the Gameblazers card example, tapping a pack creates anticipation, revealing cards one at a time repeatedly resets it, and a rare result triggers animation and haptics. The information could have appeared all at once; staging converts one result into several cycles of uncertainty and resolution.

A streak says: protect what you already own. A variable reward says: discover what comes next. The surface action may look similar, but one relies on loss aversion and the other on anticipation. The latter can still become exploitative, particularly when money and chance meet; the video presents the motivational distinction more clearly than its ethical limit.

Completion drive and the Apple Watch

The Apple Watch rings reduce the system to three unfinished shapes: move, exercise, and stand. A nearly closed circle invokes the Gestalt principle of closure. The gap remains perceptually active until the person completes it.

The video’s headline result is a 49.5% behavior change across 160,000 people, and it also reports an association between regular ring closure and lower rates of poor sleep. The important design point is not the precise number, which the video does not source in its description. The completion loop points outside the app at movement, exercise, and standing. Closing it is good only because the encoded activity can be good.

Competence instead of badge theatre

The final pattern asks whether the system shows that the person became better at the real thing. A cited meta-analysis found gains in autonomy and relatedness but little effect on competence, the psychological need most closely tied to durable intrinsic motivation.

Peloton provides live output, marks personal records, and makes a “100 rides” badge evidence of one hundred rides. Chess.com ratings model playing strength. Garmin’s training-readiness and body-battery scores try to model recovery and capacity. Each can be imperfect, but they at least refer to ability, practice, or condition rather than the number of times an app opened.

Takeaways

  • Design the game around the behavior that should actually happen.
  • Prefer local, winnable comparison over inflated global ranking.
  • Use completion loops where finishing the loop is good for the user.
  • Give streaks agency, repair, and humane escape hatches.
  • Reward evidence of skill, consistency, and real-world effort.
  • Cut mechanics that make users manage the system instead of doing the thing.

Related: dopamine-and-product-usage, variable-belohnung-und-habit-loops-in-sozialen-apps, Emotional design, Gestalt principles.

Caveat

The video presents many studies, user counts, legal developments, and effect sizes without linking them in the description. The mechanisms are useful hypotheses; the reported numbers need verification before they become evidence in design decisions.

19 paragraphs923 words6,154 characters