As festive online shopping surges, how are e-commerce firms using AI to detect fake orders, prevent fraudulent refunds? | Mint
As festive online shopping surges, how are e-commerce firms using AI to detect fake orders, prevent fraudulent refunds?
India’s festive e-commerce rush is expected to bring more orders, more shoppers and a growing fraud challenge. As suspicious transactions can resemble genuine festive buying, platforms and logistics firms are increasingly turning to AI-driven checks to protect customers without slowing deliveries.
The festive season is a crucial period for e-commerce majors and retailers to boost sales. Flash deals, discounts and last-minute purchases drive a flood of online orders, giving shoppers more choice and the convenience of faster delivery.
However, during this time, fake orders also surge, along with account misuse and fraudulent refund claims. For e-commerce platforms and logistics companies, the challenge is to identify suspicious orders, fake identities and fraudulent refund requests without disrupting the experience of genuine customers.
According to data from Datum Intelligence, India’s festive e-commerce order volumes are expected to grow from 760 million in 2025 to 966 million in 2026, marking a 27.1% increase. As more orders move through checkout, payment, delivery and returns systems in a short period, companies face greater difficulty in separating fraudulent activity from genuine customer demand.
Also Read|Lost money in an online fraud? Here’s how much you can recover under RBI’s rules
The pressure is further heightened as the nature of online shopping changes. A Datum Intelligence report published in September expects the number of festive shoppers to rise by 14.5% this season, while each shopper is projected to place 11% more orders. The average order value, however, is expected to decline from ₹1,578 to ₹1,554. This means platforms and sellers will need to process more transactions while keeping delivery and fraud-control costs in check.
Why festive sales make fraud harder to detect
Beyond transaction fraud, e-commerce businesses face several other risks, including suspicious cash-on-delivery (COD) order bookings, fake customer identities, abuse of offers, account takeovers, fraudulent return requests and delivery tampering.
These actions become harder to spot during festive sales because a transaction that seems unusual at another time of the year may appear genuine during a promotion. For instance, if a new user places several high-value COD orders, it may raise a red flag for an e-commerce platform but could also be part of normal festive shopping behaviour.
Also Read|Mint Explainer: What’s at stake for consumer companies this festive season?
The need for artificial intelligence (AI) to detect fraud is highlighted by recent industry research. According to TransUnion’s H1 2026 Top Fraud Trendsreport, 7.1% of attempted digital transactions involving consumers in India in 2025 were suspected to be fraudulent, compared with a global average of 3.8%. This highlights the importance of moving beyond static rule-based checks towards adaptive fraud-detection systems capable of identifying unusual behavioural and transactional patterns.
Traditional rule-based fraud systems typically flag transactions that meet specific criteria, such as orders above a certain value or repeated purchases from the same account or payment method. However, these checks may not be enough during a festive rush, when genuine customers may also place multiple or high-value orders.
COD and returns add another layer of risk
Logistics networks face elevated digital-fraud risks. TransUnionfound that 16.3% of attempted digital transactions involving consumers in India’s logistics sector were suspected of fraud in 2025, the highest rate among the sectors it analyzed. The volume of suspected fraud attempts in the sector also rose 31% year on year.
Cash-on-delivery orders can be vulnerable to fake bookings, incorrect addresses, delivery refusals and return-to-origin shipments. However, a failed delivery does not necessarily indicate fraud. Connected data and AI-led risk checks can help businesses identify potentially risky patterns early while limiting unnecessary RTO costs for genuine customers.
Also Read|India’s suspected digital fraud rate soars to twice the global average
Fraud risks can continue after delivery. Fraudsters may misuse return policies, make false refund claims, or use multiple accounts and devices to repeatedly exploit offers. By linking order histories, product identifiers, warehouse records, and manual review processes, e-commerce companies can help flag suspicious return or refund patterns and prevent the abuse of festive promotions.
How AI can improve fraud detection during festive sales
E-commerce platforms can use AI to identify suspicious transactions and patterns that may indicate a higher risk of fraud. One way is through automated risk scoring, which helps businesses separate potentially high-risk orders for manual review.
Using machine learning, platforms can assess a combination of factors, including order history, account details, order value, device information, delivery experience and shipping addresses. A transaction may appear legitimate on its own but could be flagged when viewed alongside related patterns across multiple orders, logistics platform Shipmaxx suggested.
However, this kind of analysis becomes difficult when customer, order, payment and delivery data are spread across multiple platforms and service providers.
Krishna Gupta, CEO of logistics platform ShipMaxx360, said, “While e-commerce brands are trying to drive sales in the festive season, this increases risks related to fake orders, COD fraud, and order returns. This is further complicated by the fact that data relating to customers, orders, payments, and delivery may be spread across multiple platforms and service providers.”
Gupta said this lack of integration makes it difficult to detect anomalies, meaning fraud may be discovered only at a later stage. He added, “AI can help analyze purchase patterns, order history, delivery patterns and refund requests to detect early risk. However, this relies on integration between different parts of your business.”
The TransUnionreport found that suspected fraud was highest at the account-login stage in India, at 3.9% of transactions. This was followed by account creation at 3.1% and financial transactions at 1.2%, suggesting that fraud monitoring needs to extend beyond payment verification.
About the Author
Swati Gandhi is a digital journalist with over four years of experience, specialising in international and geopolitical issues. Her work focuses on foreign policy, global power shifts, and the political and economic forces shaping international relations, with a particular emphasis on how global developments affect India. She approaches journalism with a strong belief in context-driven reporting, aiming to break down complex global events into clear, accessible narratives for a wide readership.<br><br>
Previously, Swati has worked at Business Standard, where she covered a range of beats including national affairs, politics, and business. This diverse newsroom experience helped her build a strong grounding in reporting, while also strengthening her ability to work across both breaking news and in-depth explanatory stories. Covering multiple beats early in her career has helped her be informed about her current work, allowing her to connect domestic developments with wider international trends.<br><br>
At Live Mint, she focuses on international and geopolitical issues through a business and economic lens, examining how global political developments, foreign policy decisions, and power shifts impact markets, industries, and India’s strategic and economic interests.<br><br>
She holds a Bachelor’s degree in English (Honours) from the University of Delhi and a Master’s degree in Journalism and Mass Communication from Guru Gobind Singh Indraprastha University. Her academic training has shaped her emphasis on precision, analytical rigour, and clarity in writing. Her interests include global political economy and the intersection of geopolitics with business.<br><br>
Outside work, Swati focuses on exploring her passion and love for food. From fancy cafes to street spots, Swati explores food like a true foodie.
E-CommerceArtificial IntelligenceFestive Season
Get Latest real-time updates
Catch all theBusiness News,Market News,Breaking NewsEvents andLatest NewsUpdates on Live Mint. Download TheMint News Appto get Daily Market Updates.
HomeAIArtificial IntelligenceAs festive online shopping surges, how are e-commerce firms using AI to detect fake orders, prevent fraudulent refunds?
More
