Marketing technology built for direct-to-consumer retail goes blind the moment it lands in India’s intermediated economy — and the failure is structural, not a matter of configuration.
Business-to-business-to-consumer (B2B2C) is not a niche pattern in India; it is the default architecture of the largest consumer-facing sectors. The pharmaceutical brand does not sell to the patient — the doctor prescribes and the pharmacy dispenses. The spirits brand cannot advertise on broadcast and does not own the transaction — the licensed outlet and the on-trade venue do. The fast-moving consumer goods (FMCG) manufacturer sells to a distributor, who sells to a wholesaler, who sells to a kirana store, who sells — in cash, unrecorded — to the consumer. In every case the intermediary owns the relationship with the end consumer, and the brand’s commercial outcome is achieved only through the intermediary’s behaviour. Western marketing stacks assume the brand can observe the chain from impression to purchase to repeat. In India, the chain goes dark at the intermediary, and everything downstream of that point — attribution, segmentation, personalisation, real-time decisioning — is starved of the individual-level data it was architected to consume.
Across a decade of deployments the pattern repeats with almost mechanical reliability. In the mid-2010s, the global enterprise stacks arrived in India carrying analytics and personalisation propositions premised on consumer transaction data. At a multi-category Indian conglomerate’s foods business, roughly ninety-five percent of volume moved through general and modern trade where no consumer-level data existed at all; the honest discovery finding was that before any of the promised analytics could run, the first-party data asset would have to be built from nothing. At a multinational pharmaceutical company’s India business, a design workshop produced the single most clarifying sentence in the entire archive of that era: pharma is indirect selling — all the other stakeholders should be touched. The team said it as a pharma insight. It is in fact the structural truth of the whole Indian consumer economy, and there was, and largely still is, no Western playbook for it.
What a decade of practice has produced instead is a three-part playbook, and it is worth stating plainly because each part inverts a direct-to-consumer instinct.
First: intermediary intelligence before consumer engagement. The instinct trained by Western martech is to start with the consumer journey. In an indirect market the intermediary is the actual commercial partner — the prescriber, the distributor, the retailer — and the first analytical investment should be understanding that party’s decision process, economics and pain points. The pharmaceutical version of this was archetype work on prescribers: differentiated engagement designed around how different doctor types actually decide, measured through prescription-audit movement rather than imaginary patient funnels. The FMCG version is the trade-cluster adaptation of behavioural segmentation: apply the clustering discipline not to consumers you cannot see but to the distributors and retailers you can — segmented by ordering behaviour, promotional response and category mix, producing actions the existing trade sales force can execute this quarter. Notably, that adaptation was never productised by the global vendors. It remains one of the most underserved analytical needs in Indian FMCG, which makes it an opportunity rather than a lament.
Second: proxy signals as interim measurement, honestly labelled. When the consumer transaction is invisible, the disciplined move is to instrument the best available proxies — social conversation as a signal of brand perception and intent, community engagement, distributor offtake as a trailing indicator of consumption — while being explicit about their biases. Social listening over-represents the young, urban and digitally active and under-represents the kirana-buying household that carries FMCG volume. Distributor data shows what moved into the channel, not what consumers chose. The era’s better engagements sized these investments accordingly: small proofs of concept, not full-stack commitments, because a proxy justifies a pilot, not a platform. That sizing discipline transfers directly to 2026, when the same proxy logic is being resold with an artificial intelligence (AI) label.
Third — and this is the one that still gets valued wrongly in Indian boardrooms: owned channels are data assets before they are revenue channels. The clearest early case came from the most constrained category. A major spirits company, prohibited from broadcast advertising, built lifestyle community platforms and an owned commerce property not because they would ever rival the licensed trade on volume, but because they created the only first-party layer in which the brand could see, segment and engage actual consumers — a relationship layer coexisting with the intermediary channel rather than replacing it. The same logic explains the direct-to-consumer (D2C) storefronts India’s large FMCG houses built from 2019 onward: measured as revenue channels against kirana volumes they are rounding errors, and finance teams periodically propose killing them on that basis. Measured as data assets — the brand’s only source of individual transaction truth, feeding segmentation and media efficiency across the ninety-five percent of volume that flows through trade — they are cheap at the price. If your D2C business case is denominated in gross merchandise value, it is denominated in the wrong currency.
The structure is loosening at the edges, which changes tactics but not the playbook. The Goods and Services Tax (GST) regime made distributor-level transactions visible as a side effect of compliance. Business-to-business (B2B) trade platforms digitised a slice of kirana purchasing. Retail media networks on the large marketplaces and quick-commerce platforms now sell brands campaign-level consumer attribution — which is the intermediary renting its data back to you, useful and worth buying, but not ownership. The intermediary still holds the relationship. The playbook holds.
On Monday, three moves. Commission an intermediary intelligence baseline before the next consumer-facing technology purchase: who are our prescribers, distributors, retailers as segments, and what does each cluster respond to — if your analytics partner cannot cluster the trade, find one who can. Re-audit every measurement claim in your current stack against the question “which of these numbers is a proxy, and what is its known bias?” — and re-label the dashboards accordingly. And re-state the business case for every owned channel you operate in data-yield terms — identified consumers added, event streams captured, media efficiency gained — before your next budget cycle does it for you in revenue terms and reaches the wrong conclusion.