Who Should Own Payment Data?
Card networks and banks quietly sell what you buy. The data has real value, the question worth asking is why none of it reaches the person who generated it.
Your card statement is one of the most revealing documents about you that exists. Where you go, when, how often, what you’re treating yourself to, what you’ve stopped affording. It is worth real money to a lot of people, and it is routinely monetised by parties you have no relationship with. None of the proceeds reach you.
What a transaction log actually reveals
Consider what a year of card transactions establishes about someone, without a single line of inference beyond the obvious.
Home neighbourhood and workplace, from where the small purchases cluster on weekdays. Commute, from the timestamps. Health, from the pharmacy and clinic entries. Relationship status, from the change in restaurant party sizes and the appearance of a second regular grocery run. Religion, from what’s closed on which day. Politics, from the donations. Financial stress, from the shift in grocery retailer and the timing of transfers relative to payday. Whether you’re travelling, and where, in real time.
You didn’t choose to disclose any of it. It’s the exhaust of buying things.
Who buys your payment data
The transaction touches at least four businesses on its way through, and several of them have found that the data is worth more than the fee.
Card networks operate analytics divisions. Both Visa and Mastercard have publicly described data and services as growth businesses offering advertisers measurement products that answer “did people who saw this campaign subsequently spend money in these stores”, built on aggregated spending across their networks.
Banks monetise their own customers’ transactions, commonly by placing card-linked offers in the banking app and taking a cut of resulting sales, and by licensing aggregated spending data.
Payment processors and fintechs sit on rich merchant-side data and sell benchmarking and analytics products derived from it.
Data brokers and hedge funds buy aggregated card-panel data to estimate a retailer’s quarterly revenue before it’s announced. This is a well-established trade with a straightforward purpose.
Almost all of this is legal, disclosed somewhere in a privacy notice, and invisible in practice.
“De-identified” carries less weight than it sounds
The standard defence is that data is aggregated and stripped of identity, so no individual is exposed.
The problem is that spending patterns are close to a fingerprint. The most-cited work here is a 2015 study in Science by de Montjoye and colleagues, examining three months of credit card records for 1.1 million people. It found that four spatiotemporal points were enough to uniquely re-identify 90% of individuals. Knowing roughly where and roughly when someone made four purchases information a friend, a colleague or an Instagram post could supply is sufficient to pick one person out of a million.
That result doesn’t mean every anonymised dataset is trivially re-identifiable, and aggregation genuinely helps. But it does mean “de-identified” describes a spectrum rather than a guarantee, and that the guarantee is weakest precisely for the kind of data being discussed here.
The bargain nobody offered you
There’s a reasonable defence of ad-supported services generally: you get the product free, the advertiser gets your attention, the platform brokers it. Crude, but at least it’s a recognisable trade.
Payment data isn’t that. You’re not getting the payment free you’re paying for it, through merchant fees embedded in prices, through account fees, through interest. The data monetisation is additional revenue on a service you already funded.
And unlike advertising, there’s no visible consideration. Nobody said “we’ll reduce your fees in exchange for your spending history”. The data simply accrued to whoever was positioned to record it, which is an accident of architecture rather than an agreement.
Why none of the money reaches you
If the data has value and its being bought and sold why does none of it reach the person who generated it?
Not for any principled reason. Because of where the data physically sits. It’s recorded by intermediaries in the course of processing, so it ends up in their systems, so it’s theirs to monetise. Ownership followed custody. Nobody decided this; it’s just what the plumbing produced.
Which suggests the fix is architectural rather than regulatory. Change where the data sits and the question of who profits from it answers itself.
What data sovereignty would actually require
Privacy and sovereignty aren’t the same ask. Privacy says don’t look. Sovereignty says it’s mine, and I decide the terms including the terms on which I’d happily share it.
That matters, because most people are not absolutists. Plenty would trade spending data for a genuinely good discount. What they object to is the trade happening without them, and the value landing elsewhere.
For that to work, three things have to be true:
The data has to sit with you. In a self-custodial wallet, your transaction history is on your device and on a public ledger. There’s no central store for anyone to license, because there’s no centre.
Matching has to happen locally. A merchant wanting to reach people who buy coffee on weekday mornings doesn’t need your history to find you. They can publish the criteria; software on your device evaluates them and decides whether to show you anything. The targeting works and the data never moves which is exactly what an on-device agent is for.
The payment has to reach you. If a merchant is willing to pay for access to your attention, that payment should land in your wallet rather than a platform’s. This is the part that needs a payment rail capable of moving very small amounts for near-zero cost, which is why this is a payments problem rather than a privacy-policy problem.
What we're buildingA non-custodial wallet where you can choose to monetise your data. Intermediaries have dominatied the advertisment scene for decades, the DYNK platform pays you directly everytime your attention is sold. Wrapped up in an agent, that works for you whilst your data stays locally on your device. Your attention should pay you not the middleman. Dynk is prelaunch, waitling is live now!
→The parts that are genuinely hard
This idea is easy to state and difficult to build, and it’s worth being upfront about why.
Public ledgers are public. On-chain transactions are permanently visible to anyone, and chain analysis is a mature industry. That’s not privacy, it’s a different distribution of exposure. Chain analysis is a mature industry, and a wallet address linked to your identity once is linked forever. Self-custody solves who profits from the data; it does not by itself solve who can see it.
Individual data isn’t worth much individually. The value in transaction data is largely in aggregate. A single person’s spending history is worth cents to an advertiser, not dollars. “Get paid for your data” pitches routinely overstate what an individual share looks like, and the honest number is small.
Markets need liquidity. A data marketplace with a handful of participants doesn’t function. This only works at scale, which is a genuine cold-start problem rather than a technicality.
Consent fatigue is real. A system that asks permission for every micro-transaction becomes noise, and noise gets clicked through. Making consent meaningful without making it exhausting is an unsolved interface problem.
Why it’s still the right direction
None of those objections change the underlying asymmetry. A person generates data of real commercial value simply by living, and every cent of it accrues to intermediaries not because they earned it, but because they hold the pen.
The economics of that only hold while the data has to pass through a central party. Once it doesn’t, once settlement is peer-to-peer, the history sits on your device, and matching happens locally and the entire justification evaporates.
You are currently the product. The interesting question isn’t how to stop being sold. It’s what happens when you’re the one setting the price.
Frequently asked questions
Do banks and card networks sell my transaction data?
Not usually as raw records with your name attached. The common practice is to license aggregated, de-identified or pseudonymised transaction data to advertisers, retailers, hedge funds and analytics firms for measuring whether an ad campaign produced real sales, or for tracking a company's revenue before it reports. Both Visa and Mastercard have publicly disclosed data and analytics as business lines.
Is anonymised transaction data really anonymous?
Weakly. A widely cited 2015 study in Science by de Montjoye and colleagues found that four spatiotemporal points were enough to uniquely identify 90% of individuals in a dataset of 1.1 million people's credit card records. Transaction patterns are highly distinctive, so de-identification is much less protective for this kind of data than the term suggests.
Can I opt out of my payment data being used?
In some jurisdictions, partially. GDPR in the EU and UK, and laws such as the CCPA in California, give rights to access, delete and object to certain processing. In practice the opt-outs are hard to find, apply unevenly across the many parties handling a transaction, and rarely cover aggregated data that has been de-identified.
What is data sovereignty?
Data sovereignty, in this context, means the person who generates data controls whether it is shared, with whom, and on what terms including being compensated for it. It goes further than privacy, which is only about restricting access. Sovereignty treats the data as an asset belonging to the individual rather than a byproduct belonging to whoever recorded it.