Gift Card Sellers: How Raise Cut Chargebacks 45%

Q: Will Sardine cut my chargebacks by 45%?

No such promise exists. 45% is Raise's reported fall over three years on its own traffic; your result depends on your fraud mix, volume and controls.

The short answer: gift card marketplace Raise cut its chargeback rate from 0.53% to 0.29% over three years with fraud platform Sardine, while manual fraud reviews fell from as much as 18% of orders to 0.25%. Those are Raise's reported results in a Sardine case study, reported by PYMNTS on October 27, 2026 — not a promise for any other merchant. The transferable lesson is structural: for gift cards, prevention at checkout beats representment after the sale, because the value is usually spent before the dispute lands.

Raise's reported results

PYMNTS reported on October 27, 2026 that gift card exchange Raise cut its chargeback rate by 45%, from 0.53% to 0.29%, over three years with Sardine, and reduced the share of orders needing manual review from as much as 18% to 0.25% — a 98% fall in manual reviews. Reviews had consumed more than 10,000 hours a year and left customers waiting on approvals. The case study says Raise has processed more than $10 billion for about 6 million users across 1,000 brands. Sardine adjusted Raise's existing fraud rules and built custom models for gift card scams, so suspicious orders are flagged without pulling good ones into review.

Before the change, Raise's in-house controls could not keep up: reviews had grown to 15% to 18% of transactions, each taking 15 minutes to 24 hours. IP masking, proxies and location spoofing passed unflagged, according to the case study. "Chargebacks were becoming a significant drain on our revenue," Raise Risk Supervisor Dartanion Wells said in the case study.

Why gift cards draw fraud

Gift cards are, in the words of Forter's Doriel Abrahams quoted by PYMNTS in 2023, "the online world's version of cash." They can be purchased digitally, transferred and redeemed before a disputed payment is identified. Fraudsters buy a card with stolen payment credentials, convert it into goods or resale proceeds — and then the legitimate cardholder disputes the original charge. PYMNTS notes the pattern Raise saw: scammers bought a card, spent it, then disputed the charge. At food and beverage retailers, Abrahams said, gift cards were 5% of traffic but 50% of fraud. The economics are unforgiving: by the time a chargeback arrives, the value is gone, the goods are gone, and representment can only argue about a sale whose proceeds have already been spent.

The merchant playbook: prevention beats representment

Raise's disclosed change was tuning existing rules plus custom models for gift card schemes — Sardine's platform evaluates device fingerprints, IP and location data, proxy use, typing cadence and mouse movement — but the case study does not list which signals Raise's deployment uses. The general prevention controls below are standard practice for this category, not claims about Raise's setup. Apply them in proportion to your volume; a small seller needs the first three, not a machine-learning team.

Velocity and denomination caps: limit how many gift cards, and how much total value, one account, device or card can buy in an hour or a day. Fraud in this category is a volume game; caps break it.
Activation-code randomisation and delayed activation: never use sequential or predictable codes, and hold activation on high-risk orders until the payment has cleared and basic checks pass. A card that cannot be spent instantly is a far weaker target.
Checkout verification: check device fingerprint, IP and proxy signals, billing-address match and account age together — no single signal decides. New account plus masked IP plus an odd denomination is the combination to hold for review, not any one factor alone.
Redemption logging: log every redemption with time, device and balance, so a later dispute can be answered with evidence of use rather than argument. If the card was spent, say exactly when and where — that record is your representment.
Make refunds easier than disputes: a buyer who can get a quick refund rarely files a chargeback. Every prevented chargeback also protects your ratio, which is what monitoring programmes actually measure.

Then price the decision: run your own numbers through the Chargeback Cost Calculator and check your ratio against the Chargeback Ratio Calculator before deciding whether to fight a case or accept it — see also the Fight-or-Refund Decision Calculator.

What the case study does not prove

PYMNTS is careful about the gaps, and so should any merchant reading the headline. The case study does not say how many reviewers were redeployed, what the labour saving was worth, or the value of chargebacks avoided. It does not say whether fewer legitimate orders are declined or whether sales rose — and PYMNTS Intelligence found 47% of merchants estimate up to 5% of legitimate orders are wrongly declined. A falling chargeback rate alongside rising false declines would be a worse business, not a better one. Treat 0.53% to 0.29% as Raise's result on Raise's traffic, over three years, with a dedicated fraud platform — not as a benchmark your store should expect, and not as evidence that any particular tool will reproduce it.

FAQ

Will Sardine cut my chargebacks by 45%?

There is no such promise in the case study. 45% is Raise's reported fall over three years on its own traffic. Your result depends on your fraud mix, volume and controls.

Why not just fight gift card chargebacks?

Because the value is usually spent before the dispute lands. Representment argues about a completed sale; prevention stops the fraudulent purchase. For this category, the dispute you never receive is the only cheap one.

What is the single cheapest control for a small gift card seller?

Velocity and denomination caps plus delayed activation on high-risk orders. They cost almost nothing to implement and directly break the fraudster's volume model.