Zepto Ads: Spend Where the Dark Store Can Deliver

Zepto compressed the distance between wanting a product and holding it to roughly ten minutes. That compression changes advertising. There is no consideration window to retarget, no abandoned cart to chase for a week. The shopper searches, sees, taps and receives. Zepto ads therefore work less like brand media and more like paid shelf placement in a store that restocks every few hours. Treat them that way and the budget behaves. Treat them like your Amazon campaigns and it leaks.

The ad surfaces Zepto offers brands

Zepto’s monetisation menu keeps evolving, but the surfaces fall into two broad families. Sponsored search placements appear when a shopper types a query, putting your SKU in or above organic results for terms you bid on. Display placements are banners and tiles on the home screen, category pages and sometimes the cart, sold for reach rather than captured intent. During festive windows the platform also packages event properties, sampling slots and co-branded visibility. Formats, names and minimum commitments change often, so check the current rate card before you plan a quarter around any single placement.

How the auction and targeting broadly work

Search placements run on an auction. You bid at the keyword or product level, and the platform weighs your bid against expected relevance and likelihood of conversion. A high bid on a poorly converting SKU still loses to a moderate bid on a proven one, which is why listing quality, imagery and ratings quietly set your effective cost. Targeting is coarser than on the big marketplaces. You steer mainly by keyword, category and city rather than by granular audience segments. That coarseness is a feature, not a gap. In quick commerce the strongest signal is the search itself, typed by someone who wants the product within the hour.

Campaign structure that survives contact

Keep the structure boring. Run separate campaigns for hero SKUs and the long tail, so a slow mover can never drain the budget of your best seller. Keep search and display in separate campaigns, because they answer different questions. Search converts existing demand. Display builds recall and prompts the unplanned add. Start in the cities where your distribution runs deepest and your ratings are strongest, then expand outward. A tight campaign in three cities teaches you more in a month than a shallow one spread across twelve.

The availability-first rule

This is the rule that separates operators from media buyers. Zepto fulfils from dark stores, and your product exists store by store, not city by city. When a store runs dry, your ad can still surface, spend the click and hand the shopper either an unavailable message or a substitution toward a rival. Before scaling any campaign, reconcile spend against availability. Track days of cover at the store cluster level and throttle ads wherever cover falls below your replenishment lead time. The discipline sounds tedious. It is also the single largest saving most brands find in their first audit.

Pacing budgets across dayparts and cities

Quick commerce demand is lumpy. Morning replenishment, evening cooking, late night cravings. A budget set to spend evenly exhausts itself during low intent hours and goes dark during the peaks that matter. Watch when your category actually transacts and weight spend toward those windows. Do the same across cities. Metro auctions are crowded and expensive, while smaller cities often deliver cheaper clicks on thinner volume. Cap each city individually so one hungry metro cannot eat the national budget by noon.

Measure returns against contribution, not ROAS

The dashboard reports returns computed on gross sales. Your business runs on what remains after platform commissions, logistics deductions and promo funding. Convert every campaign readout into contribution margin per incremental order before judging it. A flattering ROAS on a deeply discounted pack can lose money on every unit shipped, while a modest ROAS on a full margin SKU quietly builds the P&L. Watch organic rank too. Sustained paid visibility on a converting SKU usually lifts its organic position, a return the campaign dashboard never credits.

How Zepto ads differ from Blinkit ads

The platforms rhyme but do not repeat. City strength differs, so the same brand can face a crowded auction on one app and an open field on the other. Placement inventory, reporting depth and tooling maturity also differ and shift quarter to quarter. The practical answer is to run both, hold creative and offers constant, and compare cost per incremental contribution rather than headline ROAS. Whichever platform delivers cheaper profitable orders in a given city earns the next increment of budget there. We run this comparison weekly inside our Zepto Account Management engagements, and the winner flips by city more often than most brands expect.

Common failure modes

• Advertising into stores with no stock, then reading the resulting low conversion as a creative problem.
• Copying Amazon keyword lists wholesale. Quick commerce queries are shorter, more generic and more urgent.
• Bidding hard on the category head term while the pack size shoppers actually want is not listed.
• Judging campaigns at festival CPCs and killing them the week before costs normalise.
• Letting one metro absorb the entire budget because it spends fastest, not because it spends best.

A weekly loop to run from Monday

Reconcile availability against spend first thing in the week and pause stores or cities below cover. Pull cost per order by city and daypart, and move budget toward the cheapest profitable pockets. Recompute contribution per order after all deductions. Note organic rank movement on advertised SKUs. Thirty minutes, once a week, done without fail. That loop beats any clever bidding trick you will read about this year.