Ad Fraud in India’s Programmatic Ecosystem — Scale, Sources & Solutions
A brand manager at a mid-sized FMCG company in Mumbai was proud of the numbers on her screen. Her latest programmatic campaign had clocked over 40 million impressions in under a week, with a click-through rate that comfortably outperformed the industry benchmark. The cost-per-click looked efficient. The reach looked massive. But when her agency ran a third-party verification audit at the end of the quarter, the picture shifted sharply. Nearly a third of those impressions had been served to bots. A significant portion of the clicks had come from click farms operating across data centres in three different states. The campaign had performed beautifully — for everyone except the brand. This scenario is not an outlier. It plays out, in various forms, across hundreds of programmatic campaigns running in India every month, quietly draining budgets that advertisers believe are working hard for them. Estimates from independent measurement firms suggest that nearly one in five programmatic rupees spent in India fails to reach a real human being — and in a market that crossed USD 2 billion in programmatic spend in 2024, the losses are anything but abstract.
The Scale of the Problem
Quantifying ad fraud is notoriously difficult, partly because its perpetrators are constantly adapting, and partly because no single party in the supply chain has a complete view of every transaction. That said, the numbers that do exist paint a sobering picture. Industry body estimates for the broader Asia-Pacific region have consistently placed India among the top three most fraud-affected programmatic markets, alongside Indonesia and Vietnam. In a market where programmatic ad spend crossed the USD 2 billion threshold in 2024, even a conservative fraud rate of 18 to 22 percent translates to hundreds of millions of dollars in wasted budget annually — money that buys impressions no real consumer ever sees.
What makes India particularly susceptible is a combination of structural and behavioural factors. On the supply side, an abundance of low-cost, long-tail publishers creates ample opportunity for fraudulent inventory to blend into legitimate auctions. On the demand side, advertisers — particularly mid-market and regional brands — continue to optimise primarily for cost-per-thousand impressions rather than outcome-based metrics, creating an incentive environment where cheap impressions win bids regardless of quality. The race to the bottom on CPMs is, in many ways, the race to the bottom on authenticity.
Primary Sources of Ad Fraud in India
Ad fraud in the Indian programmatic ecosystem does not arrive from a single source. It is, instead, a constellation of overlapping schemes that exploit different vulnerabilities at different points in the auction process.
Bot traffic and invalid traffic (IVT) remain the most pervasive category. Sophisticated bot networks — often referred to as general invalid traffic (GIVT) at the simpler end and sophisticated invalid traffic (SIVT) at the complex end — simulate human browsing behaviour to generate fraudulent impressions and clicks. India’s large pool of IP addresses and the relatively lower cost of cloud infrastructure in the region have made it an attractive geography for bot farm operators. Advertisers targeting broad audiences across tier-two and tier-three cities are disproportionately exposed, as verification coverage in these geographies has historically been thinner.
Domain spoofing is another widespread tactic. Here, a low-quality or entirely fake website misrepresents itself as a premium publisher during the bid request, causing DSPs and advertisers to pay premium rates for inventory that never appeared on the claimed domain. The adoption of ads.txt — the IAB’s authorised digital sellers standard — has helped, but implementation remains inconsistent across the Indian long tail. Publishers who have not updated or maintained their ads.txt files create gaps that fraudsters readily exploit.
App fraud deserves particular attention in the Indian context, given the country’s enormous in-app advertising market. SDK spoofing, where malicious code generates fake ad requests and clicks without any corresponding user action, is widespread in the mobile gaming and utility app segments. Ad stacking — displaying multiple ads layered on top of one another and counting impressions for all of them while only the top ad is technically visible — is another endemic tactic in mobile environments. With India’s in-app programmatic market expected to grow substantially over the next three years, app fraud represents one of the highest-risk fronts in the ecosystem.
Click injection, a fraud type native to the mobile attribution space, also bleeds into programmatic buying. Fraudulent apps intercept app installation broadcasts and fire fake click signals just before organic installs complete, claiming attribution credit for conversions that cost advertisers real money. This is particularly prevalent in performance-heavy verticals like fintech, e-commerce, and gaming, where cost-per-install campaigns operate at scale.
Made-for-advertising (MFA) sites represent a more recent and increasingly significant source of wastage. These are websites built not to serve genuine editorial audiences but to generate programmatic revenue from unsuspecting advertisers. Typically characterised by thin content, aggressive ad density, and traffic sourced from paid redirects, MFA sites are technically not fraudulent in the strictest sense — they generate real page views — but they deliver near-zero brand value and are widely considered invalid in spirit. Research published in 2024 indicated that MFA inventory accounted for a meaningful share of open web programmatic spend in South and Southeast Asia, with India featuring prominently.
The Complicity of Opacity
One of the most frustrating truths about ad fraud in India is that it thrives not just because of bad actors, but because of structural opacity that even well-intentioned participants have failed to address. The programmatic supply chain in India, as elsewhere, involves a long sequence of intermediaries — SSPs, exchanges, resellers, sub-resellers — each of whom takes a margin and, in many cases, adds limited transparency. Advertisers who believe they are buying direct access to quality inventory may, in reality, be bidding on inventory that has passed through multiple unauthorised resellers before reaching the auction.
The supply path optimisation (SPO) conversation has gained traction in India over the past two years, with major agency holding groups and independent trading desks announcing preferred supply paths with shortlisted SSP partners. This is a meaningful step. However, SPO alone does not eliminate fraud — it reduces the attack surface by removing intermediaries but does not address fraud occurring within the preferred path itself.
The programmatic supply chain in India involves a long sequence of intermediaries, each of whom takes a margin and, in many cases, adds limited transparency. Advertisers who believe they are buying direct access to quality inventory may, in reality, be bidding on inventory that has passed through multiple unauthorised resellers.
The Measurement Gap
Addressing fraud effectively requires measuring it accurately, and this is where India faces a specific challenge. Third-party verification and measurement adoption — tools offered by platforms such as IAS, DoubleVerify, and MOAT — is concentrated among large multinational advertisers and the top tier of domestic brands. The mid-market and regional advertiser segment, which collectively represents a substantial portion of India’s programmatic spend, largely operates without independent verification. When fraud is not measured, it is not managed, and budgets continue to flow to fraudulent inventory with no signal to trigger remediation.
There is also a reluctance among some publishers and intermediaries to accept independent measurement, citing concerns about discrepancies and revenue impact. This resistance, while understandable from a commercial perspective, ultimately undermines the trust infrastructure that any healthy advertising ecosystem requires. Trade bodies including the IAMAI have called for broader adoption of measurement standards, but progress has been incremental.
Solutions: What Works and What Needs to Scale
The good news is that the tools to address ad fraud meaningfully already exist. The challenge is adoption, coordination, and will.
Universal adoption of ads.txt and sellers.json would close a significant share of domain spoofing and inventory laundering overnight. Both standards are free to implement and well-documented. The Indian industry needs a coordinated push — led by major publishers and SSPs — to make non-implementation commercially untenable, similar to what has occurred in more mature markets.
Pre-bid filtering, the practice of blocking known fraudulent or low-quality inventory before the bid is placed rather than after the impression is served, has shown strong results for advertisers who have implemented it. DSPs with robust pre-bid integrations and advertisers willing to accept slightly reduced reach in exchange for meaningfully improved quality will find the trade-off worth making.
Outcome-based buying is perhaps the most structural solution available. When advertisers optimise for verified, human-delivered outcomes — completed video views, authenticated clicks, post-click actions — rather than raw impression counts, the financial incentive for fraud collapses. The programmatic ecosystem rewards the metrics it measures. If the industry collectively shifts its measurement framework, it will shift the fraud landscape.
Private marketplaces (PMPs) and curated deals offer a curated alternative to open exchange buying, with direct relationships between buyers and vetted sellers. While PMPs are not fraud-proof, they substantially reduce the surface area for the most egregious forms of supply chain fraud. The growth of commerce media and premium video PMPs in India is encouraging in this regard.
AI-driven fraud detection is becoming more capable and more accessible. Machine learning models trained on large transaction datasets can identify anomalous patterns in traffic, bidding behaviour, and attribution signals that rule-based systems miss. Several Indian ad tech companies are building domestic fraud detection capabilities, which is a positive development — though integration with global threat intelligence databases will remain important.
An Industry Imperative
Ad fraud is sometimes treated as an unavoidable cost of programmatic advertising — an acceptable leakage in an otherwise efficient system. This framing is both inaccurate and dangerous. Fraud is not random noise; it is a deliberate transfer of value from legitimate advertisers to bad actors, enabled by opacity and under-investment in verification infrastructure. In India, where programmatic is still establishing its credibility with a significant cohort of advertisers who remain cautious about digital ROI, the continued tolerance of high fraud rates carries reputational risk for the entire ecosystem.
The solutions are not exotic or prohibitively expensive. They require coordination, transparency, and a willingness among all participants — publishers, platforms, agencies, and advertisers — to prioritise quality over volume. India’s programmatic market is at an inflection point. The choices made in the next two to three years about measurement standards, supply chain hygiene, and accountability frameworks will define whether it grows into a trusted, efficient ecosystem or continues to carry a structural fraud tax that undermines its potential.
