Fraud

Fraud is intentional deception aimed at stealing goods, money or seller privileges through fake orders, stolen payment data, returns abuse and promotional manipulations.

How fraud works: schemes and mechanism

The fraud mechanism usually boils down to three steps: the fraudster gains benefit, hides traces and tries to bypass platform or courier checks. Below are typical schemes sellers encounter on Kaspi.kz:

• Payment fraud (card fraud): use of stolen bank card data to pay for orders. Often starts with "card testing" — fraudsters run hundreds of micro-authorizations to find valid cards.
• Account takeover: a fraudster gains access to a buyer's Kaspi account and places orders using saved payment methods or Kaspi Red instalments. The guide reference: Kaspi Red.
• Friendly fraud: the buyer receives the item, then disputes the payment with the bank claiming they did not receive it or the transaction was unauthorized.
• Return/refund fraud: the buyer returns a swapped or damaged item, or uses fake tracking to obtain a refund.
• Promo abuse: creating many accounts to exploit coupons, cashback and first-time discounts.
• Collusion with courier or pickup point: fake delivery, forged signatures or delivery photos.

Why fraud is critical for a seller on Kaspi.kz

The consequences of fraud go beyond direct order losses. For a seller the concrete risks are:

• Direct financial losses: the cost of goods, logistics and fees — all deducted from revenue. Example: with an average ticket of 20 000 тг and 10 000 orders per month, 1% fraud equals 100 orders = 2 000 000 тг in direct losses.
• Commissions and penalties: returns due to unregistered disputes and chargebacks are often accompanied by holdbacks from the payment system and worsened relations with Kaspi Pay.
• Reputation and blocks: an increase in disputed orders lowers seller ratings, restricts access to advertising tools and risks temporary freezes of payouts.
• Operational costs: staff time on returns, investigations and checks — measurable in working hours and diverting resources from business growth.

Typical signs of fraud on Kaspi.kz: concrete indicators

Watch for combinations of indicators rather than a single signal. Common red flags:

• Multiple low-value authorizations or declines from the same IP, device fingerprint or delivery address (card testing).
• Mismatch between billing and shipping address, unusual BIN ranges for your market or a sudden spike in foreign BINs.
• New accounts placing high-value or multiple consecutive orders; many accounts created from one IP or phone number.
• Orders with expedited delivery to unattended pickup points, or requests to change delivery method after shipment.
• Returns arriving with inconsistent packaging, missing accessories, or tracking that doesn't match logistics records.
• Repeated disputes from the same buyer claiming non-delivery after confirmed delivery.

How to detect and prevent fraud: practical measures

Layered defence is the most effective: combine automated rules, manual checks and process changes.

• Basic payment controls: require CVV/3DS where possible, enable velocity limits (orders per IP, card, device) and block high-risk BINs.
• Order scoring: implement a simple risk score using amount, new account age, shipping vs billing match, IP/geolocation and past dispute history. Hold or flag orders above a threshold for manual review.
• Verification before fulfilment: for risky orders request additional confirmation — one-time code to phone, photo of ID, or confirmation via the buyer's Kaspi account contact.
• Stricter return procedures: require photo evidence of returned item on receipt, scan serial numbers, compare with shipping photos and keep detailed media logs.
• Logistics controls: audit delivery photos, signatures and GPS logs from couriers. For pickup points, require ID and one-time codes on collection.
• Internal processes: train staff to recognise fraud patterns, keep a blacklist of suspicious IPs/phones, and document all communications and evidence for disputes with Kaspi Pay or banks.
• Platform cooperation: promptly report confirmed fraud to Kaspi support and Kaspi Pay, provide evidence, and ask for platform-side blocks or refunds where applicable.

Practical scenarios and examples of actions

Examples of how to act in common situations:

• Card testing detected: pause suspected orders, block the source IP, and contact Kaspi Pay. If testing targets many SKUs, temporarily raise authentication requirements and enable stricter 3DS rules.
• Account takeover suspected: stop fulfilment, contact the buyer via the account email/phone, request confirmation and report to Kaspi support. Preserve all logs (IP, UA, timestamps).
• Friendly fraud dispute: collect delivery proof (photo with order number, signature, courier confirmation) and submit to the bank/Kaspi Pay as evidence. Improve proof collection for future orders.
• Return with swapped item: refuse refund until a forensic check is completed; if evidence shows fraud, escalate to platform and law enforcement with media evidence.

Checklist for implementing an anti-fraud policy (actions for 30–90 days)

• 0–30 days:
  • Enable basic scoring rules: velocity limits, CVV/3DS enforcement, block known bad BINs/IPs.
  • Define manual verification flow for orders above a threshold (value, new account, mismatch criteria).
  • Start logging all delivery media (photos, signatures) and store them linked to orders.
• 30–60 days:
  • Add IP/geolocation monitoring and automated alerts for abnormal patterns.
  • Run return-testing procedures to check pickup point integrity and courier behaviour.
  • Train customer support and logistics teams to recognise fraud indicators and document incidents.
• 60–90 days:
  • Consider integrating a third-party anti-fraud service or expanding analytics for device fingerprinting.
  • Renegotiate SLAs with logistics partners, require stronger proof-of-delivery, and implement penalties for repeated breaches.
  • Establish regular reporting on fraud metrics, root-cause analysis and an escalation path with Kaspi.

Conclusion and practical tip

Fraud on marketplaces is an operational reality — not a one-off event. The most practical defence is a layered approach: combine technical controls, stricter logistics and clear processes for verification and returns. Start with simple scoring and proof collection, iterate based on incidents, and keep close cooperation with Kaspi support and Kaspi Pay to reduce both losses and reputational risk.