Paying to Be Seen in the Aisle: Is Retail Media Creating Demand?

Imagine paying for a sign outside your shop, then paying again to be visible to someone already standing in the aisle.

That is one way to understand the uneasy attraction of retail media. Advertising close to a purchase can be useful precisely because the customer is already shopping. The same proximity makes it difficult to know how much of the resulting sale the advertisement created.

For an Indian consumer brand, the central question is not whether marketplace advertising can produce attributed sales. It is whether the next rupee buys additional profitable demand, protects useful visibility or merely takes credit for a customer who already intended to buy.

Those are different commercial jobs. A responsible retail-media strategy should say which one it is performing.

What is retail media?

For this article, retail media means advertising using a retailer's shopping environment or customer relationships. A simple example is a sponsored product placement inside a marketplace. Other arrangements can extend beyond the retailer's own site, but the details depend on the platform and contract.

Amazon India's seller advertising guide describes Sponsored Products as ads for individual product listings and Sponsored Brands as a way to highlight a brand and product collection. These are product descriptions from the seller of the advertising, not independent evidence of campaign profitability.

Bain and Flipkart's April 2026 report estimates that e-retail advertising accounted for roughly a quarter of Indian digital advertising spend in 2025. That estimate describes market allocation. It does not establish that a quarter of your budget belongs there.

The strategic attraction is understandable: a brand can communicate at a moment when the customer is comparing products. The strategic risk is that the platform can become both the shop and the place where brands must pay to be noticed.

Being near the transaction helps—and confuses measurement

A shopper searching for a category may encounter an advertised product and choose it over an alternative. That is a plausible source of additional sales. Another shopper may search for a specific brand, click its sponsored placement and purchase the item they wanted all along.

Both orders can appear in attributed advertising reports. Their incremental value may be different.

These examples illustrate a measurement problem, not a claim that branded advertising is always wasteful. Paying to maintain visibility can be a deliberate defensive choice. The brand should name it as such and examine the cost.

Attribution asks whether a sale followed a defined advertising interaction within a reporting rule. Incrementality asks what would have happened without the advertising. The first can often be reported directly; the second requires comparison and assumptions.

A dashboard cannot make that counterfactual disappear by displaying the revenue in a larger font.

Start with the retail fundamentals

Before buying more visibility, review the product page, price, availability and customer experience. An unclear listing or unsuitable pack can waste attention from even a well-targeted placement.

Ask whether the image shows the correct variant, whether the title answers the relevant question and whether the customer can understand what will arrive. Ensure that claims have evidence and that the business can honour the stated terms.

Then examine operational readiness. If the product is unavailable in the intended service area, additional exposure may have little value. If returns indicate a recurring misunderstanding, advertising can amplify the problem.

Our analysis of quick-commerce strategy explores this connection between visibility and local availability. Retail media is one part of the shelf, not a replacement for the product being on it.

Define what you are buying

A campaign might aim to win new category shoppers, introduce a variant, support a seasonal offer or maintain visibility around branded searches. Each requires a different review.

For category acquisition, examine whether the audience and product are suitable and whether new customers return at acceptable contribution. For a launch, assess whether the product earns sustained demand after the promotion ends.

For branded visibility, look at the combined paid and organic picture where the data allows it. Do not assume that every paid branded order is incremental, or that every such placement can be removed without consequence.

Set the hypothesis in plain language. “We think this pack can profitably reach shoppers looking for this use case” is more useful than “we want to optimise the funnel”. The former can be investigated.

It also makes disagreements easier to resolve. A sales team seeking turnover and a finance team seeking contribution may both be right about their numbers while discussing different objectives.

Work backwards from contribution

The advertised selling price is not the amount available to fund acquisition. Deduct the actual product, fulfilment and commercial costs that apply to the order. Include discounts and returns appropriately.

Then identify what remains before advertising and management costs. That contribution is the starting point for evaluating how much you can afford to spend, subject to your wider business objectives and accounting treatment.

Suppose a hypothetical product sells for ₹1,000 and leaves ₹250 before advertising and campaign management. A ₹300 advertising cost per order would consume that amount and more. A healthy-looking revenue multiple would not change the arithmetic.

Those figures are invented examples, not category benchmarks. Your contractual costs may differ, and repeat purchases should be included only through observed evidence and explicit assumptions.

A product with a low initial contribution might still justify an acquisition investment. State the expected payback, the risks and the evidence needed to continue. Hope should appear as an assumption, not as an invisible cell in the formula.

Ask difficult questions about the reporting window

Which interactions qualify for attribution? How long after an ad can an order count? Does the report include only the advertised product or a wider set? How are cancellations and returns treated?

Read the platform's current definitions for the format you use. Reporting rules differ and can change. Avoid comparing two platforms as though identically named metrics necessarily use the same calculation.

Keep a copy of the definitions used in the review. If the platform changes its attribution window, a jump in reported sales might partly reflect measurement rather than customer behaviour.

These questions do not imply that the reports are useless. They make the reports usable for the decision at hand.

Consider the decision threshold before choosing a measurement method. A small exploratory budget and a large annual commitment need different levels of confidence. Write down how much uncertainty the business can tolerate and what evidence would change the decision. This prevents a preliminary result from quietly becoming a permanent spending rule. It also gives the team permission to say that the evidence remains inconclusive and that the next useful step is a narrower experiment.

Test incrementality proportionately

A large brand may have the resources for a controlled holdout or geographic experiment. A small seller may need a simpler comparison. The method should match the scale and the confidence required.

Where possible, vary a defined campaign while keeping other important conditions reasonably stable. Record stock, pricing, promotions and operational changes. Compare total sales and contribution, not just the advertising dashboard.

Time-based comparisons can be affected by seasonality and competitor behaviour. Area-based comparisons can be affected by different customers and serviceability. State those limitations instead of pretending the experiment occurred in a vacuum.

If the data cannot support a precise causal estimate, report a range of plausible interpretations. A decision can be cautious without a fabricated decimal point.

The purpose is to improve resource allocation. You do not need to prove the universal effectiveness of retail media to decide whether one product and campaign deserve another test.

Avoid becoming dependent on the rented shelf

Marketplace demand can be valuable while concentrating risk. A change in placement, cost, availability or commercial terms can alter the brand's access to customers.

Maintain accurate product information and customer understanding outside one marketplace. This might include a useful brand website, authorised retailer relationships and consent-based communication where appropriate.

Do not misuse platform customer data or assume that a transaction grants unrestricted permission to contact someone elsewhere. The practical aim is a resilient business relationship, built within the applicable terms and permissions.

A brand should also know why customers choose it. If the only answer is that it bought the first visible slot, its negotiating position may be fragile.

Distinctive packaging, product suitability and dependable service can support demand across channels. They do not guarantee immunity from advertising costs, but they give the business more than one lever.

A review meeting that connects the numbers

Bring the marketplace operator, advertising manager and finance owner together. Use the same period and product set. Reconcile attributed sales with settlements and returns before debating strategy.

For each campaign, state the job, full cost, available contribution evidence and measurement limitations. Decide whether to continue, narrow, pause or investigate.

Keep a separate record of learning. A campaign that identifies an unsuitable pack or a misunderstood feature can produce a useful insight even if it does not deserve further spend. That insight should change the listing or assortment.

Our marketing-cost guide offers a wider budget framework. Retail-media spending should enter the same business review as production, management and the systems that fulfil demand.

What might happen as the channel grows?

One scenario is that retail-media offerings become easier to buy and compare. Another is that expanding formats make measurement more fragmented. A third is rising competitive pressure for the most desirable shopping moments.

These are scenarios, not forecasts of prices or market shares. Their shared implication is to preserve clear objectives, current metric definitions and independent contribution records.

Brands that know their economics can respond more deliberately to new formats. Brands that know only attributed revenue may find every new placement persuasive until the settlement arrives.

Pay for a clear commercial purpose

The best retail-media question is small enough to answer: what did this spending add, for this product, under these conditions?

Sometimes the answer will justify expansion. Sometimes it will reveal a product-page problem, an inventory constraint or expensive defence of existing demand. All are useful findings if the business acts on them.

The Umbrella can help connect brand presentation and acquisition measurement. If your marketplace reports look healthier than your margins, bring the actual settlement and campaign records. The conversation should begin with what the business keeps, not just what the dashboard counts.

Sources & Further Reading

• Amazon India: Seller advertising guide. Official format descriptions; commercial claims are the platform's own.
• Bain and Flipkart: How India Shops Online 2026. Source for the estimated 2025 share of digital advertising allocated to e-retail ads.