Series 1 — Diagnosis · Paper 4
The Chargeback Trap
When chargebacks were introduced in the 1970s, they were a genuinely good idea. Credit cards were new, consumer trust in them was fragile, and people needed a safety net — some assurance that if they handed over their card details and something went wrong, they wouldn’t simply lose their money with no recourse. The chargeback mechanism provided that assurance. It was a reasonable response to a real problem.
Fifty years later, the mechanism that was designed to build trust between merchants and their customers is quietly destroying it. Not through bad intentions. Not through fraud, at least not primarily. But through a structural design that made sense in 1974 and makes progressively less sense with every passing year — and which the card networks have had every reason to leave exactly as it is.
What a chargeback actually is
A chargeback is not a refund. That distinction matters more than most people realise.
When a merchant issues a refund, they are in control of the process. They decide whether the claim is valid. They initiate the return of funds. The relationship between merchant and customer remains intact — a problem arose, it was resolved, the transaction is closed.
A chargeback is something else entirely. When a customer disputes a charge with their bank, the money is removed from the merchant’s account before the matter is investigated — often before the merchant even knows there is a dispute. The merchant then has a defined window, typically twenty to forty-five days, to submit evidence proving the dispute is unwarranted. That evidence is reviewed not by a neutral party, but by the issuing bank — the same institution whose primary obligation is to its own account holder, the customer making the claim.
If the merchant misses the deadline, they lose automatically. If they submit evidence and the bank sides with its customer — which it frequently does, because that is what banks do — the merchant loses the transaction amount, the chargeback fee levied by the network, and whatever costs they incurred in preparing their response. The entire process can take up to a hundred and twenty days to resolve. The merchant’s money is gone on day one.
This is the mechanism that was sold to merchants as consumer protection.
The customer isn’t the problem
Before going further, something needs to be said plainly: the customer filing a chargeback is rarely acting in bad faith.
Research consistently shows that the majority of disputed transactions — estimates range from sixty to seventy-five percent — involve what the industry calls “friendly fraud”: legitimate transactions disputed by consumers who either don’t recognise the charge on their statement, forgot they made the purchase, found it easier to call their bank than to call the merchant, or simply didn’t know there was another option.
That last point deserves to sit for a moment. More than half of consumers who file chargebacks do so without contacting the merchant first — not because they want to harm the merchant, but because the banking app on their phone made disputing the charge easier than finding a phone number or writing an email. The card network designed a process that routes disputes through itself rather than back to the merchant-customer relationship. Customers followed the path of least resistance, as people do.
The result is a system that has trained an entire generation of consumers to treat their bank as the first port of call when something goes wrong with a purchase — not the business they bought from. The merchant-customer conversation that might have resolved the issue in five minutes is bypassed entirely. The customer feels processed rather than helped. The merchant feels accused rather than consulted. And the bank collects a fee for adjudicating a dispute that arguably never needed to involve it.
Nobody designed this outcome deliberately. But nobody with a stake in the current arrangement has any incentive to change it.
The economics of the chargeback, honestly stated
The card network charges the merchant a percentage of every transaction for access to its network. When a dispute arises, the network charges an additional fee — typically between fifteen and twenty-five pounds per case — to administer the chargeback process. If the merchant wants to contest the chargeback, they bear the cost of doing so in staff time, evidence gathering, and administrative overhead. If they lose, they pay again. If they win, they recover the transaction amount but not the time and cost they spent fighting for it.
The card network collects its percentage fee regardless of the outcome. The issuing bank collects its chargeback administration fee regardless of the outcome. The only party that can lose money in a chargeback dispute — or gain nothing from winning one — is the merchant.
There is a further consequence that rarely appears in the headline numbers. If a merchant’s chargeback rate exceeds a threshold set by the card scheme — currently around one percent of transactions — they enter a monitoring programme. This can result in higher processing fees, mandatory reserves held against future chargebacks, restrictions on the types of payment they can accept, or in severe cases, termination of their merchant account. A merchant can be punished for receiving too many chargebacks even when those chargebacks are demonstrably fraudulent, even when they are winning the majority of disputes they contest. The threshold measures volume, not validity.
This is not a quirk of the system. It is the system.
What this does to the merchant-customer relationship
The consequences of this structure extend well beyond the individual disputed transaction.
A merchant who has been through a chargeback dispute — particularly one they lost despite having clear evidence of a legitimate sale — learns something about the relationship between themselves and their customers. They learn that a customer can, at any moment, effectively reverse a transaction without speaking to them, and that when this happens, the merchant’s voice in the matter is limited and their likelihood of success is poor. That knowledge changes how merchants think about the customers they serve. It introduces a wariness that shouldn’t need to exist.
Customers, meanwhile, lose access to what should be the most direct and effective resolution path: talking to the merchant directly. A customer who goes straight to their bank with a dispute may get their money back, but they have bypassed the conversation that might have given them something more — an explanation, a replacement, an apology, a relationship with a business that actually cares about making things right. The chargeback mechanism treats every dispute as a financial transaction to be reversed rather than a customer experience to be repaired.
Both parties are worse off. The merchant carries a liability they have limited ability to manage. The customer receives a financial resolution but loses the human one. The bank profits from mediating a relationship it has no stake in maintaining.
The wrong instrument for the job
Here is the deeper problem, and it is one that no amount of tinkering with the existing mechanism will solve.
A chargeback is a financial reversal instrument. It was designed to answer one question: should this money go back to the customer? That is a narrow question, and the chargeback answers it narrowly — by removing the funds and adjudicating the claim through the customer’s bank.
A dispute is something entirely different. A dispute is a breakdown in a relationship between two parties who had a transaction. It involves questions that a financial reversal instrument is not equipped to ask: What went wrong? Could it have been avoided? What would actually make this right? Who bears responsibility, and in what proportion? What outcome would restore trust rather than simply settle an account balance?
These are the questions that a proper dispute resolution process asks. And the payments industry has spent fifty years using a financial reversal instrument to avoid asking them.
This is not a criticism that can be addressed by updating reason codes, tightening deadlines, or adding authentication layers. The card networks have tried all of these things. The chargeback rate is rising, not falling. Friendly fraud is increasing, not decreasing. Merchant satisfaction with the dispute process is, by any measure, poor. The instrument is not failing because it is being administered badly. It is failing because it is the wrong instrument for the job it has been given.
What a modern dispute standard looks like
If the payments industry were designing a dispute process from scratch today — with fifty years of evidence about what the chargeback mechanism actually produces — it would look nothing like the current system. The broad shape of what it would look like is not complicated. It is, in fact, obvious to anyone who has thought about dispute resolution in any other context.
It would start with direct resolution. Before any formal process begins, the merchant and customer would have a structured opportunity to resolve the matter between themselves — with the payment system facilitating that conversation rather than bypassing it. The majority of disputes that currently consume weeks of administrative effort would be resolved in this stage, because the majority of them involve misunderstandings that a single conversation would clear up.
Where direct resolution fails, it would escalate to a formal process in which both parties have an equal voice. The merchant would submit their account of the transaction. The customer would submit theirs. Both would be heard before any decision is made. Neither party would lose by default because they missed a deadline measured in days.
The adjudication would be conducted by a party with no financial interest in the outcome — not the customer’s bank, which has an obvious relationship to protect, and not the merchant’s processor, which has its own commercial interests. A neutral body, operating under defined rules that both parties agreed to when they joined the payment system, would make the decision based on evidence rather than institutional loyalty.
The outcome would be measured not just by whether funds were returned, but by whether the process was fair. A dispute rate would be understood as a signal about transaction quality and communication clarity, not just a risk metric to be managed below an arbitrary threshold.
This is not a radical vision. It is how dispute resolution works in virtually every other commercial context — from consumer arbitration to professional services to regulated financial products. The payments industry is the outlier. The chargeback is the anomaly. A proper dispute resolution framework is simply what any well-designed payment system would have built from the start.
Why this matters now
The chargeback problem is getting worse, not better. The shift to online commerce has created an environment where customers and merchants rarely interact before, during, or after a transaction. Card statement descriptions are frequently confusing — the trading name on a receipt often differs from the legal entity that appears on a bank statement, leading consumers to dispute charges they themselves authorised. Mobile banking apps have reduced the friction of filing a dispute to a few taps, further lowering the bar for claims that a single conversation would have resolved.
These trends are structural, not cyclical. They will not reverse as commerce continues to move online. A mechanism designed for the physical world of the 1970s — where the bank manager knew the local merchant, where a disputed transaction typically involved a face-to-face relationship — is increasingly poorly suited to a world where neither party has ever met and the entire transaction happened in seconds on a smartphone.
The card networks know this. They have introduced new frameworks, new reason codes, new authentication requirements. Each update adds complexity. None of them addresses the fundamental question: is the chargeback the right instrument at all?
The next generation of payment infrastructure — being designed now, on open rails that didn’t exist in 1974 — has the opportunity to answer that question correctly. Not by building a better chargeback. By replacing it with something that was designed for the job.
Thomas Larsen is a cloud platform architect and engineering leader with twenty years’ experience building open infrastructure for public-sector and defence organisations. He is currently working on an open payment scheme for the UK market.