A common point of purchase, or CPP, is the merchant, website or processor that a group of fraud-hit cards all used shortly before the fraud began. Banks and card networks find it by taking a set of cards with confirmed fraud, looking back through their earlier transactions and spotting a business that appears far more often than chance would explain. That shared location is the likely place the card data was stolen, and identifying it lets banks block or reissue every other card that passed through it before criminals use them. CPP analysis is how many data breaches, from supermarket chains to small online shops, are discovered in the first place.

Key takeaways

  • A CPP is the business that fraud-hit cards have in common before the fraud started, pointing to where the data was taken.
  • Issuing banks, card networks and specialist vendors all run CPP analysis, comparing fraud cards against normal spending patterns.
  • Most breached merchants learn about the problem from their bank or the card network, not from their own systems.
  • Once a CPP is confirmed, Visa's rules give the business 3 calendar days to report and tight deadlines for a forensic investigation.
  • If your bank replaces your card without any fraud on it, it is often because the card was used at a suspected or confirmed CPP.

Why do banks need CPP analysis?

When card data is stolen, the theft and the fraud usually happen in different places, days or weeks apart. Criminals collect numbers from one business, sell them on underground markets, and buyers then use them elsewhere, often in another country. A bank looking at a single fraudulent purchase sees only where the card was misused, not where it was compromised.

CPP analysis connects those two points. As fraud analytics firm SAS explains, it identifies the likely merchant location where card numbers were stolen so banks can stop further fraud on other cards from the same source. Because spotting a CPP depends on fraud having already happened, the faster and more automated the analysis is, the more losses it prevents.

The stakes are high on both sides. Each compromised card the bank fails to identify is a future fraud loss and an unhappy customer. Each card it reissues unnecessarily costs money and annoys a customer who did nothing wrong. Good CPP analysis aims to find the breach early and draw the boundary of affected cards as precisely as possible.

How does CPP analysis work, step by step?

The method is simple to describe and hard to do well at scale, because it means comparing a small set of fraud cases against millions of normal transactions.

  1. Build a fraud set. The analyst gathers cards with confirmed fraud over a recent period, often from chargebacks, customer disputes and fraud alerts.
  2. Pull transaction histories. For each card, they look back over a window before the first fraud, typically 30 to 180 days depending on the fraud type.
  3. Count shared merchants. Every merchant, website, ATM and processor in those histories is counted by how many fraud cards visited it.
  4. Compare with normal behavior. A big supermarket will appear in many histories simply because many people shop there, so each count is compared with how often the same merchant appears in the histories of cards that had no fraud.
  5. Rank the outliers. Businesses that appear in a far higher share of fraud cards than expected become suspected CPPs, along with the dates when the shared visits happened.
  6. Confirm and widen. The suspected window of exposure is used to find every other card that visited the same place in that period, and the finding is shared with the card networks.

The comparison in step 4 is what separates a useful CPP from noise. If 30 percent of a bank's fraud cards used a local petrol station, that matters only if far fewer than 30 percent of its healthy cards did. Analysts also look at timing: a genuine compromise tends to show fraud cards clustered within a defined window, with cards used before and after that window staying clean.

A worked example

Consider a hypothetical regional bank that sees a rise in counterfeit and online fraud on 240 of its debit cards over three weeks. Its analysts pull 90 days of history for each card and find that 168 of them, 70 percent, were used at the same family restaurant chain in the six weeks before the fraud began. Among the bank's cards with no fraud, only 3 percent had visited that chain in the same period.

That gap is the signal. The bank flags the restaurant chain as a suspected CPP, sets a window of exposure from the earliest to the latest shared visit, and finds 4,100 other cards that were used there in the window. It blocks or reissues those cards, reports the finding to the card networks, and the networks check whether other banks' fraud points to the same chain. If it does, the restaurant's acquiring bank is notified and the forensic process begins. Most of the at-risk customers never see a fraudulent charge, because their card was replaced first.

Who runs CPP analysis?

Several parties run CPP analysis, and each sees a different slice of the picture.

WhoWhat they seeWhy it matters
Issuing banksTheir own cardholders' fraud and spendingFast to act, but a small bank may lack enough fraud cases to spot a pattern
Card networks (Visa, Mastercard and others)Fraud reports and transactions across many issuersCan confirm a CPP that no single bank could see, and coordinate the response
Processors and acquirersActivity across their merchant portfoliosCan spot a compromised processor or software provider affecting many merchants
Fraud analytics vendorsPooled data from participating banksSpeed up detection for smaller institutions
Law enforcementSeized card data and criminal marketplacesCan work backwards from stolen card listings to the source

The network view is the most powerful, because a compromise at a small online store might produce only a handful of fraud cases at each of dozens of banks. Individually those cases look random; combined, the common merchant stands out. That is also how a compromise at a payment processor or a point-of-sale software provider gets traced, since the shared factor is not one shop but a service used by many.

What happens to a business identified as a CPP?

For the business at the center of a CPP, the process is formal and moves quickly. Visa's What To Do If Compromised rules, now at version 10.0 and effective from 25 June 2026, set out the deadlines.

Visa's guidance is just as firm about what not to do. A business that suspects a compromise should preserve evidence, including logs, disk images and any malware, isolate affected systems from the network rather than switching them off, and avoid logging in, changing passwords or continuing to take payments on a compromised machine. Well-meaning clean-up often destroys the evidence investigators need to find out how the attackers got in and which cards were exposed.

After the investigation, card networks can pass the costs of fraud and card reissuance back through the acquiring bank to the breached business under their account data compromise programs. How much depends on the number of cards exposed, the business's compliance status at the time and whether it cooperated. This is a strong financial reason to reduce the card data a business holds, and to run regular card data discovery scans so there is less to lose.

From the 2009 Visa malware alerts to today's indicators

Card networks do not wait for a CPP to be confirmed before sharing warnings. In early 2009, shortly after the Heartland and Hannaford breaches, Visa issued alerts to the industry listing malicious software and internet addresses linked to card data theft, so that merchants and processors could check their own networks for the same signs. It was an early example of what is now called threat intelligence sharing.

That practice is now written into the rules. Under Visa's current requirements, when Visa notifies a member of malicious IP addresses, the member has 3 calendar days to confirm an outbound firewall block, check its logs for connections to those addresses, scan its network, document what it did and submit an incident report. The same indicators that help one compromised business are used to stop the next one.

Where things stand in November 2026

As of November 2026, CPP analysis has shifted with the fraud it tracks. The move to chip cards made it much harder to turn stolen stripe data into counterfeit cards, so in-store compromises now produce less usable data. As SAS notes, fraudsters have turned to e-commerce and other card-not-present transactions instead. Online compromises, often caused by malicious scripts injected into checkout pages, are now a major source of CPPs, and they can be harder to trace because victims may have shopped at a site only once.

The rules have also tightened. Visa's What To Do If Compromised guidance was updated to version 10.0 in June 2026, and PCI DSS 4.0.1 requires businesses to protect payment pages against unauthorized script changes, aiming to stop the most common online skimming attacks before they create a CPP. Tokenized payments help too: when a contactless or mobile wallet payment uses a one-time code instead of the real card number, there is far less for a thief to steal.

What should cardholders and businesses do?

If your bank replaces your card unexpectedly, accept it, even if you have seen no fraud. It usually means your card passed through a suspected compromise point and the bank is acting before criminals do. Check recent statements for small test transactions, which fraudsters often use to confirm a stolen card works, and report anything you do not recognize.

Businesses should assume they could become a CPP and prepare. Keep card data out of your systems where possible through hosted payment pages, tokenization and point-to-point encryption, make sure your logging would show what happened if you were breached, and know who to call at your acquiring bank. A business that responds within the deadlines, preserves evidence and cooperates fully is in a much stronger position than one that finds out from its bank and is caught unprepared.

Frequently asked questions

What does CPP mean in credit card fraud?

CPP stands for common point of purchase: the merchant, website or processor that a group of cards with fraud all used before the fraud started. It points to where the card data was most likely stolen.

How do banks know where my card was compromised?

They compare your card's recent transactions with those of other cards that suffered fraud. If many of them used the same business during the same period, far more often than normal, that business becomes the suspected source.

Why did my bank send me a new card when there was no fraud?

Banks often reissue cards that were used at a suspected or confirmed compromise point, as a precaution. Replacing the card before criminals use the stolen data prevents fraud rather than reacting to it.

Can a small business be identified as a CPP?

Yes. Card networks combine fraud data from many banks, so even a small online shop can be identified if cards with fraud across several issuers share it as a common merchant.

What happens if my business is a common point of purchase?

Your acquiring bank or the card network will contact you, and you will need to report, preserve evidence and usually appoint a PCI Forensic Investigator within strict deadlines. You may also face costs for fraud and card reissuance, depending on the investigation's findings.