Nagarika is a game you play by walking your own street. This page sets out why the game needs a reward system at all, what each reward idea is, and what is still open. Updated 6 October 2026 after review.
Open Show details on any section for the full working.
500 points does not mean ₹500. Pride XP is a score inside the game and nothing else. It buys no cash, no goods and no discount. What a point is eventually worth, if anything, is not defined and is not being proposed here.
Every number on this page is a placeholder. 500, 200, 100 — these were set to get a playable build into the field. They are held as configuration and can be changed in minutes. The numbers are to be finalised together; what needs agreement now is the shape of each reward, not its figure.
What is here. Why a reward system is needed, what is running today, the reward ideas on the table, and the questions still open.
The Hosur Civic Awareness Program genuinely works. Its limit is physical: it is capped by how many volunteers can be present, and engagement fades once a drive ends. Nothing carries the behaviour through the week after.
That gap is the whole reason a reward system exists. It is not points for entertainment. It is the thing that has to do what a volunteer standing beside someone does — give a reason to act on a day when nobody is watching and nothing is organised.
Every reward on this page is judged against that one line. The test applied to each is simple: is this a useful action? If the answer is no, it earns nothing, however easy it would be to measure.
Indians manage private space well and externalise public space — not my problem, somebody else's job. The same person abroad is a model citizen. The behaviour is not missing; the social contract is. A reward system is one way to supply the missing half of that contract: a visible, immediate acknowledgement that acting on your own street counts for something.
Bottom-up alone collapses: children change, parents do not, children give up. Top-down alone is too slow. Both have to fire together until the behaviour is unmissable in one geography.
There have been many apps that let people complain. If five of the six ways to earn points are complaining, this becomes one more complaint registry — and complaint registries do not change behaviour.
That is why the proactive mechanisms matter more than their point values suggest. What is needed is not only more points for proactive behaviour, but more ways to earn through it, so that a resident's ordinary week is mostly doing rather than mostly reporting.
The behaviour actually wanted is prevention — a resident who segregates, follows the collection schedule, and does not litter in the first place. That produces the absence of a bad act, which cannot be observed. So every reward has to pay for an observable proxy instead, and choosing proxies is the whole of the problem.
These narrow the field before any specific idea is considered. Two do most of the work: reward impact, not persistence rules out anything that pays more the longer a problem sits, and pride first sets a ceiling on how large a prize can be before it becomes the reason to take part.
| Principle | What it means |
|---|---|
| Points for doing, not knowing | Every point requires a verifiable real-world action. The game cannot be won from a screen. |
| Every earning must be a useful action | If repeating it adds nothing in the real world, it earns nothing. |
| Reward impact, not persistence | Value is tied to problems being resolved quickly, never to problems staying open. |
| Democratise participation | Reporting a problem and registering a complaint need no level, no permission and nothing earned first. |
| First mover rewarded most | The first person to surface a problem earns the most; later reports earn sharply less. |
| Speed is the pressure | The faster a problem is resolved, the larger the reward that survives to be shared. |
| Pride first | Recognition is the primary engine. Material rewards stay a light supplement. |
| Carrot, positive framing | Messaging states the benefit created. Never fear, never penalty. |
| Usable by anyone | The benchmark is WhatsApp. If a step is less intuitive than that, it is wrong — people who are uncomfortable with technology have to be able to play. |
| Action | Points | Note |
|---|---|---|
| Report a black spot first reporter | 500 | A large dump. Reporting means photographing it — there is no other way to report |
| Report a flashpoint first reporter | 200 | A small dump |
| Report one already reported | falls sharply | Second reporter roughly half, third a tenth, and down from there. Still counted, barely rewarded |
| Verify a spot is clean | 120 | Closes the spot and credits the people who drove it |
| Remind the authority | 40 | "This spot is still not gone." Useful every time, unlike re-photographing it |
| Daily handover — segregated | 300 | Once a day. The one action that pays for your own waste |
| Daily handover — mixed | 100 | Once a day |
| Starting balance | 890 | Every account, so nobody opens at zero |
Segregated pays three times mixed. That ratio matters more than any other number here, because every reporting action rewards finding rubbish — which on its own rewards going to look for more of it.
The first person to surface a problem has done something genuinely useful. The fiftieth has not. The award falls on a logarithmic curve — roughly half for the second reporter, a tenth for the third, and on down — so there is no way to farm a single spot and no reason to let a problem accumulate reports instead of being fixed.
Repeat reports are still counted. The count is evidence of how many people a problem affects, which is what gives an escalation weight.
It used to be a separate earning line. It is not one: reporting a spot is photographing it, so paying twice for the same act was double counting. Walking past the same spot every day and re-photographing it adds nothing in the real world either.
What does add something is pressure. So the repeat action that earns is a reminder to the authority, which is useful every time it is sent.
Logging when a spot was reported and when it was verified clean produces a record of how long each one took. Over a few months that is a scoreboard for the Corporation itself — this black spot took 25 days to clear — built from residents' own observations rather than from anyone's opinion.
Twenty spots in Ward 23 were surveyed on foot on 30 July 2026 with photographs and GPS coordinates. Nothing on the map is invented.
A resident photographs the waste they are handing over and marks whether it is segregated. Segregated pays 300, mixed 100, once a day.
It is deliberately not verified. Confirming a bag really is segregated needs the waste-type classifier, still in training, and waiting for it would have meant shipping nothing. Points are paid for the action and the photograph is kept, so every submission is a labelled example — the unverified version is what makes a verified one possible.
When a spot is accepted, it opens with a bounty. The bounty falls every day the spot stays open. Whatever is left when the spot is verified clean is shared out among the people who worked on it.
The direction is the point. A bounty that grew would reward leaving a spot to rot. Falling means the community's attention is worth most when the problem is fresh — which is also when it is cheapest to fix.
| Cleared on | Bounty surviving | Effect |
|---|---|---|
| Day 1 | 10,000 | A large reward shared by the people who pushed it |
| One week | ~4,000 | Still worth having |
| One month | ~2,500 | Sharply reduced |
| Two months + | 250 | Neglected — a nominal amount only |
Who gets it. The top three contributors on that spot, ranked by the points they earned on it. Reporting, reminding and verifying all count toward that ranking, so no separate calculation is needed — the points a person already earned on the spot decide their share.
Earlier the thought was eight to ten names. On a phone, tapping a cleared spot and being shown ten names reads as nothing at all. Three names are read. The cut-off also makes the ranking honest, because it is decided by points already earned rather than by anyone's judgement of who "contributed extensively".
A bounty for clearing a spot is a reason to create one. Two people could dump rubbish, report it, clear it and split the reward.
This cannot be designed out completely, and pretending otherwise would be worse than naming it. Three things reduce it:
The working assumption is guarded trust rather than blind trust. Roughly ten per cent of people will do the right thing whatever happens, ten per cent the wrong thing whatever happens, and the eighty per cent in between follow whatever the system appears to expect of them. The design is aimed at that eighty.
The idea the rest of the design leans on: make the reward social rather than material. A resident's name stays attached to the spots they helped reclaim, and standings are published on a regular cadence.
| Element | What it is | Status |
|---|---|---|
| Name on the reclaimed spot | Tap a cleared spot and see how it got cleaned, and the top three people by points earned on it | Agreed |
| Leaderboard | Individuals ranked by points | Running |
| Zone board | Zone against zone within the ward | Running |
| Level | Your step-by-step progress. Levels come quickly, so there is always something near | Running |
| Rank | A title covering a band of levels — Nagarik, Sevak, Guardian, Rakshak, Ward Champion. Level is a subset of rank | Running |
| Achievement categories | A separate, parallel track: one achievement per topic you contribute to | Idea |
| Fortnightly report | A published at-a-glance summary of the boards, which can be printed and put up | Idea |
A level is the small step; a rank is the band of levels it sits in. Levels move quickly so a resident feels progress in the first week. Ranks are slower and are the thing worth saying out loud. The alternative — five ranks and nothing else, where the first one takes four months — gives a new player nothing to see.
Not a replacement for rank and not a label for the kind of person someone is. One achievement per topic, earned by crossing a points threshold within it.
Today there is one topic, garbage. If traffic is added later, a resident active in both holds both achievements. Nobody is sorted into a group — the same person can be a Guardian by rank and hold a garbage achievement and a traffic achievement at once.
Names and thresholds are not settled, and the earlier set — Green Guardian, Infra Defender, Social Warrior, Culture Keeper, Technocratist — was a first sketch rather than a proposal.
Everything is computed in real time: a resident opens the app and sees exactly where they stand. The fortnightly piece is separate — a short compiled report of the leaderboard and zone board that can be circulated or printed and put up on a board, so the standings reach people who are not looking at the app that day.
These carry more weight locally than a prize of the same nominal value, because they come from an institution rather than from us.
Periodic sponsor-backed prizes for leading contributors, kept deliberately light so that recognition, not the prize, remains the reason to take part. No sponsor has been approached and no prize has been agreed.
They bring partners into the programme. A sponsor who funds a prize has a reason to stay involved, and the prize is a recruitment tool in a ward where nobody has heard of Nagarika.
Small, frequent and recognition-adjacent rather than large and rare: a book, a plant, a reusable segregation bin — something that is itself part of the behaviour. Never cash or anything with resale value.
The idea was that a zone performing well could raise a request — an upgraded bus stop, say — and the programme would help push it with the Corporation. Whether this is feasible has not been checked with them.
This is the part that decides whether Nagarika is a civic game or another complaint app. Reporting, reminding and verifying are all reactive. Only one proactive mechanism is running today, and one against five is the wrong balance.
What is needed is more ways to earn through proactive behaviour, not simply more points on the one that exists.
| Mechanism | What it rewards | Status |
|---|---|---|
| Daily handover | Photographing the waste you hand the sanitary worker, marked segregated or mixed | Running |
| Restraint log | Recording an occasion you did not do the wrong thing | Agreed |
| Daily tasks | Small assigned checks — "this spot was cleared yesterday, is it still clean?" | Idea |
| Least-maintenance street bounty | The street that needs the least work from the collector earns for everyone on it | Open |
| Civic worker rating | The collector rates each street; residents rate the collector | Open |
A resident records an occasion on which they did not do the wrong thing: put a wrapper in their pocket and brought it home rather than dropping it. A small number of points, once a day.
It can be faked, and that is accepted deliberately. The award is small and capped at one a day, so a dishonest player gains almost nothing. The purpose is not the points — it is that someone who has logged it twenty times starts to notice what they are claiming. Awareness is the mechanism. It is the only idea here that works through the player rather than on them, which is exactly why it must never be worth enough to be worth gaming.
Self-reported points are shown on a separate line and kept off the competitive leaderboard, so honesty-based points never cast doubt on measured ones.
Rather than leaving a resident to work out what to do, the app can assign a few small things: this spot was cleared yesterday, walk past and confirm it is still clean. Tap it and the map takes you there.
This is what turns maintenance into a rewardable action without paying people to re-photograph the same rubbish. It is also how a daily or weekly streak would work, and it is the simplest way to add proactive earning without inventing a new mechanic.
Everyone living on a street earns when that street needs the least work to keep clean. It does two jobs at once: it rewards prevention, which nothing else does, and it removes the incentive to manufacture a problem, because the street loses more than the individual gains.
It needs a measure, and that is the open part — see the next item.
While collecting, a worker grades each street on a simple scale — one to five — for how much work it took to keep clean. A street where there is almost nothing to do scores differently from one that takes real effort. That produces a per-street cleanliness score from day one, which nothing else in the system produces, and without which the street bounty cannot be built.
The worker should earn nothing for the rating. If a worker is paid for rating and residents are paid when the rating is good, nobody has any reason to report a dirty street. Paying nothing removes the conflict and makes an honest rating the path of least resistance.
To balance it, residents rate the worker and the worker rates the street — two directions, so neither side alone decides the record.
Subjectivity remains: a worker who has fallen out with someone on a street could mark it down. Reducing that without destroying what makes the rating valuable is an open problem.
Shorter than it was. The reporting curve, the bounty, the top-three rule, the spot definitions and the removal of duplicate earning lines were all settled in review. What remains is listed below.
| Question | Where it stands |
|---|---|
| 1 · How is the street measured? | The least-maintenance bounty needs a per-street signal. The collector's rating is the only candidate that works from day one. Asking residents to self-report their own street does not, because everyone on it gains from the same answer. |
| 2 · How is the collector's rating kept honest? | Paying the worker nothing for it removes the obvious conflict, and a two-way rating balances it. Reducing the remaining subjectivity is unresolved. |
| 3 · How is gaming detected? | A watchdog role plus analytics that flag the patterns. Needs someone to own it and a definition of what gets flagged. |
| 4 · What else counts as proactive? | Four mechanisms exist; only one is running. More ways to earn through doing rather than reporting is the single most valuable thing to add before launch. |
| 5 · Any material reward in the pilot? | Recognition only is the view offered. Not decided. |
| 6 · Is the zone civic improvement feasible, and should it be offered? | Feasibility has not been checked with the Corporation, and the entitlement framing has to be resolved first. |
| 7 · The numbers. | Every figure is a placeholder, to be set together. These are configuration and can change at any point without a rebuild. |