All articles Fraud & ATO

Virtual Card Free Trial Abuse: How to Catch It

One glowing blue card panel echoed by ghosted duplicates above laptop and phone silhouettes

A prospect signs up for your trial on Monday with a card ending 4417. On Wednesday another signup arrives: different email, different issuer, a card minted two minutes earlier. That is virtual card free trial abuse in its usual shape — an app issues a fresh number in seconds, your processor returns a valid authorization, and your product hands out another trial.

It is hard to catch because each system sees one slice. The processor sees a card. Your email service sees a mailbox. Analytics sees a cookie that got cleared, or an incognito window. Nothing in that stack sees the person, and your funnel is built around account identity, which is cheap to regenerate.

So stop asking whether the card is bad and start asking whether you have seen this person before. Three signals answer that: card BIN data, email age, and a device-level visitor ID. Alone, each either misses the abuse or punishes honest buyers. Combined into one policy, they catch the person cycling cards while leaving the privacy-conscious buyer who uses a virtual card on purpose alone.

Read the BIN, then stop trusting it

The first six to eight digits of a card number identify the issuer and the card program: network, funding type, country of issue. It is cheap context, it usually arrives with the authorization response, and it is worth storing as a category rather than a verdict.

What the BIN gives you

A stable way to group signups by card program, which is exactly what you need for counting: how many trials this program has opened today, how many devices the same BIN has appeared on, whether the BIN has a history of disputes on your own account.

Where it runs out

One virtual card program can mint thousands of numbers that share a single BIN, so the BIN tells you “this is a virtual card from program X” and never “this is trial number four.” Treat it as context. Never block on it alone: privacy-conscious buyers use virtual cards deliberately, and they are often your most security-aware customers.

Email age and device IDs: what a card swap cannot reset

Developer at a dark desk with glowing blue tiles grouped above the keyboard

Email age, measured from data you already hold

Disposable-domain lists, the shape of the local part, and your own records of prior trials, tickets and invoices cover most repeat signups before you consider a vendor. Ask whether the address has appeared in a previous trial, whether its domain accepts mail at all, and whether the signup pattern matches a scripted flow. If you use a third-party email-age service, confirm your lawful basis and that your privacy notice covers it, and have counsel review it.

Device-level visitor IDs

The card and the inbox are disposable. The device usually is not. A visitorId from Prynt Identify stays stable across cleared cookies, incognito windows and a new card, and it arrives with the signals you need to interpret it: VPN, proxy, Tor and datacenter detection, bot and automation signals, and tampering indicators. Device fingerprinting explains what that identifier survives and where it stops being reliable.

Two counters carry most of the weight: how many accounts this device has already created, and how many devices this card has appeared on. Prynt Control tracks cards per device, devices per card and accounts per card, and keeps cards only as keyed hashes, so you get the counting without storing the number.

Turn three signals into one free-trial decision

Write the policy as a decision your code can act on, not a score someone has to interpret at 2 a.m. Prynt Control answers the free_trial use case with allow, step-up or block, plus reason codes and a suggested step such as verify_email or card_or_passkey. A workable starting default:

  • Virtual card alone: allow.
  • Virtual card, brand-new email, and a device with a prior trial: block, or step up to card or passkey verification if the account looks valuable.
  • Known device with a prior trial and no other matching signal: step up.
  • Everything else: allow and log it.
{
  "use_case": "free_trial",
  "decision": "step-up",
  "reason_codes": ["virtual_card", "new_email", "device_with_prior_trial"],
  "suggested_step": "card_or_passkey"
}

Protect the honest virtual-card buyer

Single signals should never block. A shared office laptop, a family tablet, a QA script or a buyer who uses virtual cards by habit will each produce one or two matches without any abuse. Set thresholds on counts you can defend, route anything near the line to step-up instead of a hard block, and give the customer a path to resolve it.

Phone and laptop linked by blue threads to a translucent panel with abstract fingerprint rings

Then watch before you enforce. Prynt Control opens with a 14-day mirror period in which nothing is blocked and every decision is logged, so you can see who your policy would have stopped before it stops anyone. If the pattern looks coordinated rather than like one person cycling cards, the notes in fraud rings apply, and payment fraud detection covers what changes once the trial converts to a real charge.

FAQ

Is a virtual card a fraud signal on its own?

No. It describes the payment method, not the person. Use it as one input and let device and email history decide the outcome.

How do I measure email age without a data broker?

Start with disposable-domain lists, MX lookups, local-part patterns, and your own records of prior trials, support threads and invoices. That covers most repeat signups.

What if two real people share one device?

Give them a way through: email or passkey verification, or an extra seat. A decision the customer can resolve beats a hard block on a shared laptop.

Try it free

Start on the Free plan with 5,000 identifications a month, no card required. Run the three signals in mirror mode, read the decisions and reason codes, then turn enforcement on at pryntid.com.

Two engineers reviewing abstract blue graphics on a dark wall display

Try it free

Prynt is device intelligence with a free plan: 5,000 identifications a month, visitor IDs with confidence and bot detection — no credit card. Upgrade to Pro for VPN, proxy, tampering and behavioral signals.

Keep reading