A ticketing platform's checkout page was compromised from October 8 to 11, 2026. Injected skimmer code captured the Primary Account Number (PAN), CVV, expiration date, billing ZIP, and email address as customers placed orders. The company determined on October 30 that customer information may have been affected. That's a gap of nearly three weeks between the end of the exposure window and the determination.
These dates highlight why we're addressing these questions. They're the ones payment security engineers have been asking since the disclosure. Some answers are uncomfortable.
Q1: Four days of exposure and 19 more before they knew who was hit. Where does that time actually go?
Two clocks run in a skimming incident. The first is detection: how long the script runs before anyone notices. The second is scoping: how long it takes to determine which sessions and transactions were exposed. The dates here suggest both took time, and scoping is often underestimated.
If you don't retain a record of which script versions were served to which sessions, scoping becomes forensic reconstruction instead of a database query. Log the hash and source of every script that executes on payment pages, and keep those logs long enough to answer "who loaded the bad version" without guessing. That's your scoping shortcut.
Q2: What should real-time monitoring on a checkout page actually look for?
Start with what PCI DSS v4.0 requires. Requirement 6.4.3 calls for an inventory of every script on payment pages, with each one authorized and its integrity assured. Requirement 11.6.1 requires a change- and tamper-detection mechanism that alerts on unauthorized modification of HTTP headers and payment page content as received by the consumer's browser, evaluated at least weekly or at a frequency defined by your targeted risk analysis.
Weekly is a minimum, not a goal. For a checkout page, weekly evaluation leaves room for a skimmer to run for days. Useful Indicators of Compromise (IoCs) include:
- New or changed script hashes on payment pages
- New outbound domains receiving POST requests from the browser
- Form-field listeners attached by scripts you didn't ship
- Content Security Policy (CSP) violation reports, especially for
connect-srcandscript-src
Pipe CSP reporting into your alerting, not a dashboard nobody opens. A skimmer that exfiltrates to an unlisted domain should trip a violation the moment it fires.
Q3: We already scan our servers. Doesn't that cover us?
No. Server-side file integrity monitoring won't see a script injected through a compromised third-party tag, or one that only manifests in the browser. The skimmer is assembled in the consumer's session, which is why 11.6.1 specifies content "as received by the consumer's browser." Your origin can be clean while customers are being skimmed.
You need a control that observes the page the way a customer's browser does. That means synthetic browser checks from outside your network, client-side monitoring, or both. If you reduce your payment page's script footprint to what's strictly necessary, you also reduce what has to be monitored.
Q4: What does a good response look like once we spot a skimmer?
Run it as a defined playbook, not an improvisation. The order matters:
- Contain first. Remove the malicious script or roll back to a known-good build. Where feasible, take the payment page to a hosted or redirected payment flow while you investigate.
- Preserve evidence. Capture the injected code, the injection point, and server and CDN logs before cleanup overwrites them.
- Define the window. Determine first and last execution using your script-serving logs.
- Notify. Requirement 12.10 requires an incident response plan that's ready to be activated, with defined roles, communication procedures, and contact strategies for the payment brands and your acquirer. Know those contacts before you need them.
- Find the root cause. Skimmers get in through compromised third-party scripts, vulnerable plugins, or stolen admin credentials. Closing the page without finding the entry point invites a repeat.
Contain-then-investigate is the right default. Investigating while the script is still live just extends the window.
Q5: How do third-party assessments fit in? Is an annual pen test enough?
An annual test finds what exists on the day of the test. Skimming attacks exploit change: a new tag, a vendor update, a dependency bump. Treat external assessment as a recurring input, not a yearly event.
Practically, that means scoping assessments to include the client-side attack surface, not just the application and infrastructure. Ask your assessor to enumerate every script on payment pages and trace each to an owner. Assessments should also cover your change-approval path for tag managers, since a tag manager with weak access control is an injection point. Requirement 6.4.3's script inventory and authorization is a good test: if an assessor can find a script you can't justify, you've found your gap before an attacker did.
An external assessor also brings different eyes. Internal teams stop seeing scripts they've lived with for years.
Q6: The company is offering two years of credit and dark web monitoring. Does that actually protect customers?
It helps, but understand what it doesn't do. Credit monitoring watches for new credit accounts opened in someone's name. The data exposed here (PAN, CVV, expiration date, billing ZIP, email) is card-not-present fraud material. It's used to make purchases on existing accounts, which credit monitoring doesn't catch. Dark web monitoring tells people their data has surfaced; it doesn't remove it or stop its use.
The more effective customer-side controls are the ones that make stolen Cardholder Data useless fast. One-time or merchant-locked virtual cards limit a stolen PAN to a single use or merchant. Issuer-side alerts and prompt card reissuance cut the usable life of exposed data. In the US, consumers also have dispute rights under the Fair Credit Billing Act for unauthorized charges, so statement monitoring and fast reporting to the issuer matter more than any enrollment.
For your own program, the lesson is that post-breach services are a goodwill gesture and a regulatory courtesy, not a substitute for preventing the capture. Don't count them as risk reduction in your own analysis.
Where to go for more
- PCI DSS v4.0, Requirements 6.4.3, 11.6.1, and 12.10 for the controls discussed above
- PCI SSC guidance on e-commerce security and payment page protection
- OWASP Top 10 for injection and supply-chain risk on the client side
- NIST Cybersecurity Framework (CSF) for structuring detect, respond, and recover functions around your incident playbook
Pull your current payment page script inventory this week. If you can't produce it in an hour, that's your first finding.





