Smart ticketing is one of the most widely documented technologies in this atlas — fare systems on six continents. It is also the one where the hard part is provably not technical: every documented struggle here is about migration and adoption, not about readers and cards.
This guide connects to the e-ticketing definition and covers systems from Hong Kong's 1997 card to electric BRT corridors opening in 2024.
| Generation | Examples | Defining trade-off |
|---|---|---|
| Closed-loop city card | Hong Kong 1997, Paris 2001, London 2003 | Total control, total responsibility — the city issues and runs the payment system |
| Open-loop bank card | Milan, Rome, Sofia | No card to issue; fare logic moves to the back office |
| Account-based | Singapore SimplyGo | Fares computed centrally; the token is just an identifier |
| App / super-app | Budapest, Kyiv, Berlin | Ticketing becomes one feature of a wider city relationship |
Notably, none of these generations replaced the previous one. Octopus — launched in 1997 — still reports around 15 million transactions a day and roughly 98% penetration among residents aged 15 to 64, a level of adoption no later system in this atlas has matched.
The clearest financial mechanism in this topic is not efficiency. It is money that used to disappear. Kigali's Tap&Go has carried 298 million+ journeys since 2015, and operators recover 30–40% of fare revenue previously lost to cash handling. That is the argument that funds a ticketing programme where ridership growth alone would not.
Lagos' Cowry card and Curitiba's fleet telemetry sit in the same tradition: the value is in knowing what was actually carried and actually collected.
Ljubljana's Urbana card states the recurring lesson most plainly: winning residents over to a single card was the hardest part, requiring a year-long transition and heavy public communication before legacy systems could be retired. The card worked long before the city could switch the old one off.
Dubai's nol upgrade shows the same in a live schedule: 40% complete in March 2025, 72% by July 2026, with full bank-card and wallet acceptance slipping from Q3 2026 to Q1 2027. Nothing failed. Payment migrations simply take years, and plans that assume otherwise are the thing that fails.
The one outright failure in this topic came from the layer above. Whim bundled transit, taxi, bike-share and car rental into a single subscription — and an independent evaluation found its subscribers made 73% of trips by public transport against 48% for the average Helsinki resident. The operator went bankrupt in 2024 anyway. Ticketing integration is solvable; the commercial model on top of it is not automatically so.
Sofia's system makes the case compactly: any globally issued bank card, phone wallet or QR ticket works, so no proprietary city card had to be issued at all, and a daily fare cap means three or more trips never cost more than a day pass. It processes close to a million journeys a day.
The trade-offs are real and worth naming: open-loop hands the payment rails and their fees to card networks, needs a fallback for residents without bank cards, and moves fare capping into a back office the city must still operate. Barcelona and Madrid made the same move; Istanbul and Hong Kong keep systems whose penetration would be reckless to disturb.
Tallinn is the useful outlier, and one of the few peer-reviewed entries in this topic: registered residents ride free, so the fare system's job is not collecting money but verifying residency — which also gives residents a reason to register, with the tax revenue that follows. Before optimising fare collection, it is worth asking what the fare is for.
The practical takeaway: the technology is settled and open-loop is the reasonable default for a city starting now. Plan the migration, not the launch — and be clear whether the system exists to collect fares, to measure ridership, or to give residents a reason to be registered at all.
Four transit cards that look identical at the gate and answer to four different owners. The card is the most visible piece of a city's transport data layer, and it is the piece a city most often does not control.
| Project | Since | Status | Outcome | Evidence | Run by | Cost | Who issues it, and who keeps the fare |
|---|---|---|---|---|---|---|---|
| beep Card: Metro Manila's Concession-Run Transit Fare SystemManila, 🇵🇭 | 2015 | Live | Ongoing | Official | Private company | not public | A private consortium paid the state for the concession and recoups it from riders — the card is not the transport agency's. |
| PRESTO: a provincial fare card on a city transit systemToronto, 🇨🇦 | 2012 | Live | Ongoing | Independent | Utility or public operator | 2006 Accenture master agreement $232M CAD, expanded by more than $1.7B in work added without competitive procurement (Ontario Auditor General, 2020) | The card belongs to a provincial agency; the city's transit system buys it as a managed service and does not set its terms. |
| CharlieCard and the AFC 2.0 Fare System ReplacementBoston, 🇺🇸 | 2006 | Live | Ongoing | Independent | Utility or public operator, Private company | $935.4M contract, project estimated at $960.2M, up from about $723M at award in 2017 | The transit agency issues it and keeps the value on the card itself — the design weakness that outlived two public demonstrations of it. |
| Cowry Card Contactless Transit PaymentLagos, 🇳🇬 | 2021 | Live | Ongoing | Official | Utility or public operator | not public | The transit authority made one card the only accepted fare on its own corridor, and most riders still prefer cash. |