On Flipkart Ads, the ROI figure in your dashboard is usually inflated by shoppers who were already searching for your brand or already on your listing, so it hides whether ads are creating any new sales. Split brand, category and competitor searches into separate campaigns, test whether your brand-term spend is actually needed, and move the budget it frees up into category searches for the few hero SKUs that can win a side-by-side comparison, judged against a break-even ROI worked out from your real margin.
Why Most Flipkart Ad Accounts Look Better Than They Are
Open almost any Flipkart Ads dashboard for a D2C brand that has been advertising for a few months and the headline number looks healthy. The ROI figure, which Flipkart reports as ad-attributed revenue for every rupee spent, sits comfortably above what the founder expected. The problem is that this number is an average, and the average hides where the revenue came from.
In most accounts we audit, a large share of the attributed revenue comes from three kinds of shopper who were never really in play. People who typed your brand name into the search bar. People looking at a listing of yours that already ranks at the top of its search results. People who had already viewed or added your product and came back to finish the purchase. Ads shown to these shoppers convert brilliantly, because the decision was mostly made before the ad appeared. That is exactly why they drag the account ROI up, and exactly why that ROI tells you very little about whether advertising is growing the business.
Brands coming from Amazon sometimes find the Flipkart numbers confusing, because Amazon reports ACoS, spend as a share of sales, while Flipkart reports ROI, sales as a multiple of spend. They are the same measurement turned upside down: ACoS is simply one divided by ROI. So when a founder asks how to reduce ACoS on Flipkart, the real question is how to raise ROI on the spend that matters, without inflating it by buying more of the sales you would have made anyway.
| Flipkart ads on your brand name | Flipkart ads on category searches |
|---|---|
| The shopper already searched for you by name | The shopper is still choosing between brands |
| High ROI, because most would have bought anyway | Lower ROI, but many buyers are new to you |
| Worth defending where rivals bid on your name | This is where ad spend actually creates sales |
| Tightly capped budget, reported on its own | Most of the budget, held to a break-even floor |
None of this means the brand-term spend is wasted by definition. It means it is doing a different job from the rest of the account, and as long as the two are blended into one number, you cannot tell whether either job is being done well.
Separate the Searches You Would Win Anyway
The fix starts with structure. Instead of one or two campaigns covering the whole catalogue, split spend by the intent behind the search, so that each campaign has one job, its own budget and its own target.
Brand defence. Searches containing your brand name or your product line names. Keep this campaign small, cap its budget, and report its ROI separately. Its job is to stop a competitor sitting above your own listing when someone searches for you, not to generate a flattering number.
Category growth. The generic searches a shopper types before they have picked a brand: the product type, the problem it solves, the attribute they care about. This is where most of the budget should sit, because it is the only place where ads regularly put your product in front of someone who was not already looking for it. It will show a lower ROI than brand defence. That is expected, and it is the campaign that deserves the most attention.
Competitor and adjacent searches. Searches for rival brands or neighbouring categories. This is optional and usually the least efficient, but for a challenger brand with a clear price or quality argument, a small, tightly watched budget here can win shoppers who are open to switching.
Flipkart also offers automated campaigns that manage bids towards a target ROI you set. They are useful, but understand what they optimise for. A system told to hit a return target will naturally gravitate to the cheapest conversions it can find, and the cheapest conversions are very often shoppers already searching for you or already sitting on your listing. Left alone as the main campaign, an automated setup can report a strong ROI while spending a large share of the budget on sales that needed no help. Use it as a discovery tool with a capped budget, read which queries and placements it found, and move the ones that bring in genuinely new buyers into manual campaigns where you control the bid.
If this structure sounds familiar, it is the same principle behind separating brand terms from the rest of an Amazon PPC account. The platforms differ in their formats and reporting, but the reason for the separation is identical: a blended number cannot tell you which decision is working.
Test Whether Your Brand Ads Are Actually Incremental
Separating the campaigns shows you how much you spend defending your own name. It does not tell you whether that spend is necessary. For that you need a simple test, and most brands never run one because the brand campaign looks so profitable that nobody wants to touch it.
The question the test answers is narrow: if you stopped paying for this search, would the sales still happen? A listing that already ranks first organically for its own brand search will often collect most of those orders without the sponsored slot. A listing where competitors actively bid on your name, or where a reseller or look-alike product sits just below you, may lose real sales the moment you step back.
- Pick a SKU that already ranks first For its own brand search, in an ordinary week well away from a sale event
- Record total sales, not ad sales Organic plus paid units for that SKU over a normal baseline period
- Cut brand-term bids sharply Leave category campaigns untouched so only one thing changes
- Search your brand name every day Note whether rival listings take the sponsored slots above yours
- Compare total sales and decide If they held, move the budget to category searches. If they fell, restore it
The detail that makes or breaks this test is what you measure. Ad-attributed revenue will obviously fall when you cut the ads, so it tells you nothing. What matters is total sales of that SKU across organic and paid. If total sales hold roughly steady, the brand spend was mostly buying orders you would have received for free, and that budget is better used on category searches. If total sales drop, or a competitor immediately takes the top sponsored slot on your own name, the defence was earning its keep and should come back.
Run it one SKU at a time, and never in the weeks leading into a sale event, when demand swings for reasons that have nothing to do with your bids. And keep checking the result every few months. Competitors discover your brand searches over time, and a name nobody bid on last quarter can be contested this quarter.
"The most profitable campaign in your account is often the one doing the least work. Test it before you trust it."
- Brand Integer Marketplace Growth Team
Put the Budget Where Shoppers Are Still Choosing
Once brand spend is contained, the budget it frees up needs somewhere better to go. On Flipkart that usually means category searches and the placements shoppers see while comparing products: sponsored positions in search results, and ad slots on other product pages in the same category, where a shopper looking at a rival listing can be shown yours.
Be selective about which products you push into those placements. A shopper who has not heard of you is comparing your listing against everything else on the page, so an ad only pays off if the listing can win that comparison. That means a competitive price within its band, a rating that does not put people off, complete attributes so the product shows up in the filters shoppers use, and ideally the Flipkart Assured badge. Everything that helps a listing rank organically in Flipkart search also decides how hard your paid clicks work, because the same shopper is judging the same page. Put category budget behind two or three hero SKUs that meet that bar, rather than spreading it across a catalogue where half the listings are not ready to convert a stranger.
Listing health also changes week to week. A product whose rating has slipped after a bad batch, or whose stock is running low, should come out of category campaigns until it recovers. Paying to send new shoppers to a listing that is about to disappoint them, or to go out of stock, is one of the fastest ways to turn ad spend into lasting damage. If a rating dip is the problem, deal with it at the source first; recovering from a run of negative marketplace reviews is a product and operations job before it is a marketing one.
Sale events need their own plan. Competition for placements rises sharply around Big Billion Days, click costs climb with it, and the automated campaigns that behaved sensibly all year can burn through a budget in hours. Decide in advance which SKUs get funded during the event, at what ROI floor, and with what daily caps, as part of the wider preparation for Big Billion Days and the Great Indian Festival, rather than raising budgets on the morning the sale opens.
Read the Numbers That Actually Tell You Something
A cleaner structure only helps if the weekly review looks at the right figures. Start by working out a break-even ROI for each hero SKU. Take the selling price, subtract Flipkart's fees, shipping, the product cost and the cost of returns, and what is left is the contribution margin you can afford to spend on acquiring the order. Break-even ROI is the selling price divided by that margin. A category campaign running below it is losing money on every order it drives, however good the account average looks.
Be careful with returns in particular. The revenue in the ads dashboard is counted on the orders placed, before cancellations and returns play out, so in categories where returns are common, such as fashion and footwear, the real return on ad spend is lower than the screen suggests. Set the ROI floor for those SKUs higher to compensate, and check it against settled payouts rather than attributed revenue once a month.
Then read the search queries, not just the campaign totals. Each week, look at which queries in category campaigns spent money without producing orders, and exclude the ones that are clearly irrelevant. Look at whether your brand name is creeping into category campaigns and inflating their numbers, and exclude it there so it stays in the defence campaign where it belongs. And track how your total Flipkart sales move alongside ad spend over time. If spend rises and total sales stay flat while ad-attributed sales climb, the ads are claiming credit for orders that were already coming. That single comparison is the most honest health check any Flipkart ad account has.
Frequently Asked Questions
What is a good ROI on Flipkart Ads?
There is no universal figure, because a good ROI depends entirely on your margin. Work out the contribution margin on each hero SKU after Flipkart's fees, shipping, product cost and returns, then divide the selling price by that margin to get your break-even ROI. Anything above it is profitable, anything below it loses money on each ad-driven order. Judge category campaigns against that break-even figure, and do not compare them with your brand defence campaign, which will always show a much higher ROI because those shoppers were already looking for you.
Should we bid on our own brand name on Flipkart?
Often yes, but only as a small, separately budgeted campaign, and only after checking that it is needed. If competitors or look-alike products bid on your brand name, a defence campaign stops them sitting above your listing when a shopper searches for you. If nobody does and your listing already ranks first for its own name, much of that spend may be buying orders you would have received anyway. Run a simple test on one SKU, cutting brand bids and watching total sales rather than ad-attributed sales, before deciding how much to keep.
Are Flipkart's automated campaigns worth using?
They are useful for discovery, but risky as the main campaign. An automated campaign working towards a return target will naturally find the cheapest conversions, and those are often shoppers already searching for your brand or already viewing your listing. That produces a strong ROI without necessarily growing sales. Give automated campaigns a capped budget, review which queries and placements they find, and move the ones that bring in new buyers into manual campaigns where you control the bid and can keep brand searches out.
How do we reduce ACoS on Flipkart without losing sales?
Flipkart reports ROI rather than ACoS, and ACoS is simply one divided by ROI, so the goal is to raise ROI on spend that creates sales. Start by separating brand, category and competitor campaigns so a flattering brand figure is not hiding weak category spend. Remove queries that spend without producing orders, pull listings with slipping ratings or low stock out of campaigns, and concentrate category budget on two or three hero SKUs that can win a side-by-side comparison. Cutting bids across the board lowers ACoS too, but usually by cutting the sales that ads were genuinely creating.
