If your retail media programme is still “marketplace ads, run part-time by the e-commerce manager,” you are not slightly behind best practice — you are a full maturity cycle behind what is now the ordinary practitioner standard.
The Indian side of this is immediate. A consumer brand of any scale now buys placements across eight or more retail media networks — the horizontal marketplaces, the beauty and fashion verticals, three or four quick commerce apps — each with its own auction, its own reporting, its own definition of an attributed sale. That fragmentation is an operating problem, and operating problems are not solved by media buying instinct. They are solved by discipline: standard workflows, a common measurement currency, and people whose actual job this is. India’s retail media market is young — it compressed a decade of Western development into roughly five years — which means Indian brands can see their own future by looking at how the mature markets industrialised the practice. We have spent time inside that industrialisation, and the gap between it and the average Indian brand’s “sponsored listings budget” is the point of this piece.
Start with how the mature networks themselves are built. The large United States grocery and pharmacy retail media networks are not ad products bolted onto a website; they are staffed operating organisations. A campaign there moves through a defined pipeline — commercial intake and response, annual and per-campaign planning, creative development with fixed feedback rounds, activation and ad operations with mandatory pre-launch quality gates, then billing and reconciliation — with named owners at each phase. The division of labour is consistent across networks: the retailer keeps what only the retailer can credibly hold — the commercial relationships, the first-party audience data, and above all measurement, because advertiser trust in closed-loop attribution collapses the moment it is outsourced — while specialist agency partners run the technical execution layer of planning, buying and trafficking. Service is packaged in graded tiers, with deeper audience access and faster reporting cadence at higher commitment levels. And the standard arc is agency-built, then insourced on an explicit two-to-three-year timeline. None of this resembles “ads on a marketplace.” It resembles a business unit.
Now look at the buy side, where the professionalisation is just as concrete. The large agency groups’ commerce practices train their people to a written insight standard: a performance observation is not reportable until it states what happened, why it happened, and what the client should do about it. A number without a recommendation fails review. Those practices run formal certification curricula on their retail media management platforms, maintain a defined data-source hierarchy so every figure in a client report has a known provenance, and deliver against fixed weekly, monthly and quarterly reporting cadences. That is what “practitioner standard” means: not talent, but codified method.
The clearest marker of maturity, though, is the currency. First-generation retail media reporting ran on return on ad spend (RoAS) — attributed sales divided by spend — and RoAS flatters relentlessly, because it happily counts shoppers who searched for your brand by name and would have bought anyway. The discipline has moved to incremental return on ad spend (iRoAS): the return on sales that would not have happened without the ad, estimated through test-and-control designs. iRoAS matters doubly in a fragmented market because it is attribution-agnostic — it is the one metric that can be compared across networks that each define an attributed sale differently, which is precisely India’s eight-network problem. Around it sit the rest of the working vocabulary: share of voice on the search terms that matter, tracked as a leading indicator of position; and bid discipline that prices keyword classes differently — generic category terms, your own brand terms, competitor terms — because the economics of each class are different. Practitioners run to explicit thresholds by class. The precise numbers are house-specific and proprietary; the existence of the method is the standard, and a brand whose bids make no distinction between its own name and a generic category term is signalling that no one is running the programme.
The money explains the professionalisation. The scale of the mature market — the figures below are widely published industry estimates, offered as reported orders of magnitude rather than audited fact — is what turned a side channel into a discipline:
| Market (reported industry estimates, illustrative) | Indicative scale |
|---|---|
| United States retail media spend, 2023 | Roughly USD 45 billion |
| United States retail media spend, projected toward 2028 | Roughly USD 130 billion |
Channels do not roughly triple inside five years and stay amateur. The talent, tooling and method followed the money in the West; the same sequence is now running in India, three to five years behind, and the brands that industrialise early will buy better than the ones that wait.
Here is the dual lens. The brand voice worries, fairly, that all this operating rigour optimises the bottom of the funnel and starves the work that builds pricing power. The evidence voice answers: the discipline is exactly what protects the brand from that fate. Undisciplined retail media spend drifts to defending your own brand terms — the least incremental rupee available — because it produces the prettiest RoAS. It takes iRoAS, share-of-voice tracking and keyword-class economics to prove that drift is happening and redirect the money to spend that actually acquires anyone. Rigour here is not the enemy of brand building; it is what stops retail media quietly becoming a tax on demand you already own.
On Monday, three tests. Pull your last retail media review deck and check every page against the practitioner bar: does each observation state what happened, why, and what you will do — or is it a screenshot of a dashboard? Ask your team, or your agency, for your iRoAS by network and by keyword class; if the answer is a RoAS number, or silence, you have found your maturity gap. Then make it someone’s job: a named owner for retail media across all networks, a single measurement currency, a bid policy that distinguishes brand, generic and competitor terms. The mature markets have already written this operating manual. India’s version will not take a decade to arrive — it is arriving now, and it will not wait for your org chart.