A payment network processes transactions. That processing generates a data asset as a by-product — a record of what was purchased, at what merchants, in which categories, at what values, in which geographies, at which times — that is unlike any other dataset in the world. Not because the network has access to information others do not. But because the network’s position in the middle of the ecosystem means it sees every transaction across all participants simultaneously, in near real time, without any individual participant being able to see anyone else’s data.
The network does not know who the cardholder is. It sees a tokenised card reference, a merchant, a category, an amount, a geography, and a time. The cardholder’s identity, their account details, their KYC profile — none of that is available to the network. What the network sees is behavioural: patterns of spending activity aggregated across the cardholder population, observable without personal identification.
That combination — full-population coverage, near-real-time frequency, cross-participant visibility, without individual identification — is the data asset. No bank has it. No retailer has it. No government statistical agency has it. The network has it as the structural consequence of processing the world’s card transactions.
What issuers cannot see about their own cardholders
An issuer manages its cardholder portfolio with data that is limited to transactions on its own cards. It can see what its cardholders spend at every merchant — but only on that issuer’s cards. It cannot see what those same cardholders spend on competing cards, in which categories their wallet share is split across multiple payment methods, or how their total spending compares to peer cardholders at other institutions.
The network sees all of it. The same cardholder who uses Issuer A’s card for groceries and Issuer B’s card for travel is visible to the network as a spending pattern across both cards — anonymised, without the identity connection, but present as a behavioural signal that the category spend of the cardholder population is distributed across institutions in knowable proportions.
A portfolio benchmarking product that gives each issuer an anonymised view of how their cardholders’ category spending compares to equivalent cardholder cohorts across the scheme — showing where the issuer has strong wallet share and where it is losing transactions to competing payment methods or non-card alternatives — creates insight that the issuer cannot generate from their own data. The issuer does not receive any data about individual cardholders at other institutions. They receive aggregated, anonymised comparative intelligence about category spend distribution. The distinction is what makes the product commercially and legally viable.
What merchants cannot see about their category
A merchant sees its own sales. It cannot see what its customers spend at competitors, what drives category spending decisions in its segment, or how its own performance compares to similar merchants in the same geography and demographic profile.
The network sees merchant-level transaction flows across the full scheme. It can observe — anonymised and aggregated — that a specific merchant’s share of its category’s total network spending has changed, that the spending pattern of its customer base has shifted toward or away from specific occasion types, or that competitors in the same category and geography are experiencing different volume trajectories. None of this requires identifying individual cardholders. It is pattern-level intelligence drawn from the aggregate transaction data.
A merchant analytics product that benchmarks each merchant’s performance against anonymised peer merchants creates insight that no other data provider can supply — because no other data provider sees the full category spending across all merchants simultaneously. That differentiation justifies the pricing premium that generic market research cannot.
Economic nowcasting and the timeliness advantage
Official consumer spending statistics lag actual economic activity by weeks to months in most major economies. A quarterly retail sales report published six weeks after the quarter ends reflects economic activity that payment network data recorded in near real time at the point of purchase.
The network that processes consumer card transactions sees the same spending patterns that official statistics later measure, before those statistics are published. Economic nowcasting — generating near-real-time estimates of consumer spending activity, category-level retail performance, and economic momentum from network transaction data — produces intelligence that is genuinely different from anything official statistical sources can provide.
The product requires careful design. The anonymised transaction data used for nowcasting must be processed under appropriate privacy frameworks, the methodology must be validated against official statistics to demonstrate accuracy, and the use cases must be defined to meet applicable legal requirements for data use. These are tractable challenges for an organisation operating the data governance infrastructure of a major payment network. The resulting product — real-time consumer spending intelligence for policymakers, central banks, and institutional investors — commands premium pricing because no substitute exists.
The data governance requirement
The commercial viability of network spending intelligence products depends entirely on the integrity of the privacy framework underlying them. Individual cardholder data must remain at the issuer. The network’s intelligence products must be built on anonymised aggregates that cannot be reverse-engineered to identify individuals. Regulatory compliance across all jurisdictions where the data originates adds further requirements.
These constraints are not obstacles to the data product business. They are the foundation of it. An intelligence product built on data that participants, cardholders, and regulators trust to be properly anonymised and governed is a defensible commercial asset. An intelligence product that generates privacy concerns is a liability that destroys the underlying data relationship.
What success looks like
The primary metrics are data product revenue as a proportion of total network revenue, data product gross margin, participant engagement rate with analytics products, and renewal rate. Data product revenue at major networks is growing faster than transaction fee revenue and carries significantly higher margins. The compounding nature of the relationships — issuers and merchants that use network analytics to improve their programmes attribute performance improvement to the network relationship — makes the long-run commercial value substantially higher than the initial revenue contribution suggests.
