Three Years of Palm Payment in China: A Field Read on How Biometric Payment Actually Scales
Three Years of Palm Payment in China: A Field Read on How Biometric Payment Actually Scales
TL;DR
In May 2023, Beijing's Daxing Airport Express became the first metro line in the world to accept palm pay at the gate — running on our palm recognition platform. Eighteen months later, the same technology was running in 1,500 7-Eleven stores, gyms, sports centres and — by November 2024 — its first overseas Visa pilot in Singapore. Three years in, the interesting question is no longer whether palm pay works. It is what shape its adoption curve takes — and what that shape predicts for markets now beginning their own contactless-pay journeys. The lessons travel; the conditions do not always.
The Big Picture: A Real Adoption Curve, in Public
Most discussions of biometrics payment still operate at the level of forecasts. Analysts project growth, vendors announce partnerships, consultancies publish trend decks. What has been comparatively rare is a deployment large enough, public enough, and old enough to actually study. We've now operated one long enough to look back on.
The publicly observable timeline is short and dense. Beijing's Municipal Commission of Transport confirmed the Daxing Airport Express palm-print fare gates on 21 May 2023 — our first public deployment. By September 2023, more than 1,500 7-Eleven stores in Guangdong province had switched on the same underlying platform, integrated with Weixin Pay. By the time CNN visited the rollout in January 2024, a Supermonkey gym chain in Shenzhen had also signed on, alongside corporate canteens, a sports centre, and select campus and shared-charging deployments. In November 2024, we announced a partnership with Visa at the Singapore Fintech Festival, taking palm payment outside mainland China for the first time.
Three years from launch to international pilot is a reasonably short curve for a new payment system, particularly one that requires hardware at the point of sale and a behavioural change from the consumer. The reasons it moved that fast are not random.
The First Lesson: Scenarios Diffuse in a Specific Order
Looking back, the diffusion did not start where most retail-focused commentary expected — in shops. It started at the gate.
Transit was the first domino: a captive, repeating-user audience with high tolerance for trying something new in exchange for a faster commute. The Beijing pilot was followed within months by China Daily's reporting on extension into sports centres, campuses, catering and shared-charging — environments characterised by frequent, low-value, repeat interactions where the upside of skipping a phone or wallet is felt every time.
Convenience retail came next, with the Guangdong 7-Eleven rollout. Convenience stores share transit's properties — high frequency, low ticket size, an audience that prefers speed over choice — but add a critical commercial dynamic: tap to pay or contactless pay is no longer just about checkout speed; it becomes a way to fuse identity, payment and loyalty into a single act. A returning customer who pays by palm has by definition revealed who they are. The member-card layer that retailers used to chase across separate apps now arrives with the transaction.
This pattern — transit, then high-frequency low-ticket retail, then identity-bound services — is a useful lens to bring to any new market. The order is set by user tolerance, not by vendor preference.
The Second Lesson: China's Curve Was Structurally Enabled
This is the part that travels less well, and it should be said plainly.
China's mobile-payment market is unusually concentrated — Mordor Intelligence estimates digital wallets account for around 72 per cent of payments by mode in 2025, with most of that volume sitting inside two super-apps. That matters for one reason: enrolment friction. To pay with a palm, a consumer must register both a payment instrument and a biometric template. In a market where most consumers already keep payment instruments inside a super-app they open daily, enrolment is one screen away. In markets that split daily payments across a card network, several wallets and a contactless layer, that moment is harder to engineer. Hardware density compounds the effect: a local POS supply chain was able to push palm scanners into transit gates and convenience counters quickly, in a way thinner local ecosystems cannot match.
None of this argues against palm pay outside China. It argues against assuming the time-to-1,500-stores number repeats. The shape of the curve will — the order — is the more transferable insight.
What the Daxing-to-Guangdong Curve Looks Like in Operation
Beyond the headline timeline, what makes this case worth studying is the operational detail underneath the curve. In our Guangzhou BRT (bus rapid transit) deployment, for example, the system handles intermittent connectivity by polling order status and retrying payment requests rather than failing the transaction outright — a small piece of engineering that becomes very visible when a city's morning commute depends on it. The same kind of operational thinking shows up in retail: our 7-Eleven Guangdong rollout delivered a 25 per cent peak-hour cashier-efficiency uplift, and used the device to fuse member identity with payment in a single act.
This is also the most honest place to put our own role in the broader picture. The technology underneath the Daxing pilot, the 7-Eleven rollout, and the November 2024 Visa partnership in Singapore is the same product family — what we now offer as the PayMax payment platform — and we've reached around 50 million users domestically across 100,000+ merchants and 20+ scenarios. We've declined to share more granular transaction figures publicly, which is part of why the 1,500-store 7-Eleven rollout and the Daxing Airport Express deployment have become the most-cited public anchors. The data point that makes this a case study, though, isn't the scale itself — it's that the same operator runs the transit layer, the convenience layer, and now the international card-network layer. That's what makes the order of diffusion observable in the first place.
The Third Lesson: Honest Sceptics Help, and the Industry Should Listen to Them
Adoption curves are easier to celebrate than to interrogate. It's worth pausing on the parts of the rollout that have drawn legitimate criticism — including ours.
CNN's January 2024 reporting on the rollout included a representative concern from Edward Santow, an industry professor of responsible technology at the University of Technology Sydney:
"When your personal information is hoovered up at a huge scale, that creates a kind of honeypot for cybercriminals."
That concern is real and not specific to palm pay; it applies to any large-scale biometric system. Our answer in the same CNN piece — encrypted templates, opt-in service, user-set per-transaction limits — is reasonable, but it's also exactly the kind of design choice buyers in EU, ASEAN and LATAM jurisdictions are likely to codify into procurement requirements over the next eighteen months as biometric-specific regulation matures (see, for example, the EU AI Act's biometric provisions, which began to take partial effect in 2026).
The right reading is not that palm payment is too risky to scale. It's that the markets entering their own scale phase will scale on stricter terms than China did. Consent UX, on-device template handling, deletion timelines and fallback flows for users who can't or won't enrol biometrically are now part of the procurement conversation, not a post-launch retrofit. We think that's a good thing — and we're designing accordingly.
What This Means Outside China: Different Markets, Different Curves
Our Visa pilot announced at the 2024 Singapore Fintech Festival is the first concrete look at what palm payment looks like when the market structure changes. The pilot is small — a single café (Alchemist), three banks (DBS, OCBC, UOB), a defined set of Visa cardholders, and operations run through our local Singapore subsidiary. What's notable is the shape, not the size: Singapore's biometric-pay audience is enrolled through a card network rather than a super-app, and the consumer's payment instrument is a Visa token bound to the palm template, not an internal wallet.
It's one of at least three structurally different overseas pathways now visible:
| Pathway | Who anchors the consumer relationship | Public example | What it suggests about the curve shape |
|---|---|---|---|
| International card network | Visa / Mastercard / equivalent | Our Visa Singapore pilot, 2024 | Cross-border breadth before single-market depth |
| Domestic bank or financial group | Local state or commercial bank | Pilots run through partner banks in markets where domestic payment rails dominate | Single-market depth driven by an incumbent's existing user base |
| Security / hardware OEM channel | Local access-control or device manufacturer | Public-domain reporting on Japan-market access-control integrators trialling palm | Slower payment use, faster identity / access use |
In comparative context, BiometricUpdate's coverage reads the Singapore move as a category shift: palm payment moving from a single-market story into one underwritten by global card networks, alongside Amazon, J.P. Morgan and Mastercard. Amazon's own palm pay deployment — Amazon One, which Fast Company reported reaching all 500-plus Whole Foods locations by the end of 2023 — is the comparable physical-footprint datapoint in the US, and it took roughly three years from the technology's 2020 launch to that footprint. Different market, different curve, similar inflection point.
What This Means for Decision Makers
| If you are… | Consider… | Timeline |
|---|---|---|
| A retail or convenience operator | Mapping where palm pay's transit-then-convenience pattern would meet your customer base; biometric pay tends to scale where ticket size is low, frequency is high, and identity already matters (loyalty, age verification, member access). | Pilot decision in next 6–12 months |
| A bank or card-network strategist | Watching the Singapore pilot model — palm payment via a tokenised card, enrolled at merchant POS — as a precedent for non-super-app markets. | Track quarterly through 2026 |
| A transit or public-venue operator | Studying the Daxing and Guangzhou BRT operating patterns (network-resilient retry, gate-throughput tolerance) before designing your own pilot RFP. | Design phase 6–12 months; pilot 12–18 months |
| A regulator or compliance lead | Codifying consent UX, template-handling and fallback-flow requirements before a domestic palm-payment scale phase, not after. | Now |
| A CTO or product leader at an enterprise | Asking whether the contactless pay decision is really about checkout speed, or about fusing identity with payment — these are very different procurement conversations. | Strategic review in next quarter |
Frequently Asked Questions
How big is palm payment in China today?
We've reached around 50 million users across 100,000+ merchants and 20+ scenarios — including transit, retail, gyms, sports centres, hospitals and corporate canteens.
How does palm payment work at the gate or counter?
A user enrolls once — at a station ticket machine in transit, or at the merchant's POS in retail — by associating a payment instrument with a one-time palm scan. After enrolment, the user holds their hand above the scanner; the device matches the palm against a template, applies any merchant rules (member coupons, age checks), and returns a result to the POS or gate. We use both palm print and palm vein patterns and apply data-masking and encryption to the templates.
Is palm pay safer than tap to pay or other contactless pay methods?
The honest comparison is "different attack surface, not categorically safer." Tap to pay relies on tokenised card data and a physical card or device. Palm pay relies on a biometric template that, in well-designed systems, is irreversibly transformed and never stored as a raw image. Each has a distinct failure mode: a tokenised card can be lost or transferred; a biometric template, if compromised, cannot be reissued the way a card number can. Procurement-grade evaluation increasingly looks at both categories of risk together, rather than treating biometrics as a binary upgrade.
Where We Go Next
China's curve is unusual to study because the same operator sat behind every step of it — the transit pilot, the 7-Eleven rollout, and the Visa partnership. That continuity is what makes the diffusion order observable. For organisations now beginning their own evaluation, the practical question is less "is palm pay real?" and more "which of the three pathways above maps to our market?" Our PayMax product line covers large-scale payment deployments; our public success-stories index covers operating records across healthcare, transit, retail and access control for buyers who want to read the full record before opening a conversation with us.
Related Resources
- PayMax product overview
- How 7-Eleven achieved 25% faster checkout with palm payment in China
- Visa launches palm payment in Singapore — partnership case study
- Industries Overview
About Tencent PalmAI
Tencent PalmAI is an AI-powered palm recognition service that combines palm print and palm vein identification into a single contactless biometric act.
In the context of palm payment specifically, we're the layer underneath the public deployments referenced in this article — from transit gates to convenience-store counters to the first cross-border card-network pilots — and the operating record that makes a longitudinal field read like this one possible.
