Quick Commerce Daypart Pacing: Ads by Hour and Store

Quick commerce demand is hourly, but most budgets are daily

A brand running ads on Blinkit, Zepto and Swiggy Instamart usually sets one number per campaign: a daily budget. The platform then spends that budget in whatever order impressions arrive. In quick commerce, impressions do not arrive the way they do on a marketplace. They cluster hard into three windows, and a flat budget usually empties into the cheapest, lowest intent hours of the day long before the highest intent hours arrive.

We have seen this pattern in campaign logs repeatedly. A Rs 4,000 daily budget on a keyword campaign is exhausted by 2:40 pm, with well over half of the day’s platform orders still ahead of it. The campaign reports a respectable 4.1 return on ad spend and looks healthy in the weekly deck. It is still leaking, because the hours it never bid in are the hours that convert best.

The four dayparts that actually exist

Across Indian metro dark stores the intraday order curve is stable enough to plan against. Approximate shares of daily orders look like this:

• Morning, 7 am to 11 am: 15 to 20 percent of orders. Milk, bread, eggs, atta, tea and coffee. Small baskets, high frequency, very little discretionary spend.
• Midday, 11 am to 5 pm: 20 to 25 percent. The flattest window. Household, personal care and top up shopping. The cheapest clicks of the day and the weakest intent.
• Evening, 5 pm to 11 pm: 40 to 48 percent. This window decides the month. Snacking, beverages, ready to cook, ice cream, chocolate, gifting and everything impulse led.
• Late night, 11 pm to 2 am: 8 to 12 percent nationally, meaningfully higher in Bengaluru, Gurugram and Hyderabad. Tiny baskets, very high impulse conversion, thin bid competition.

Category shifts this curve, so measure your own before you plan against the average. A breakfast cereal brand sees a sharper morning hump and a weak evening. A cold beverage brand in May sees an afternoon spike that does not exist in December. A condom or a snacking brand sees a late night share closer to 20 percent. Pull four weeks of hourly order data from the platform reports, or reconstruct it from your own sales feed, and plot it once. That single chart is worth more than another month of bid tinkering.

Emulating dayparting when the console does not give it to you

Most brand teams do not have clean hourly scheduling on every ad product they buy. The workaround is structural, not clever. Build three campaigns for the same SKU set: Morning, Midday and Evening. Keep the keyword lists identical so the data stays comparable. Then run them with manual gates.

The operating pattern that works: Morning campaign live from 6:45 am with roughly 20 percent of the daily budget, paused at 11 am. Midday campaign live with 15 percent, deliberately underfunded, paused at 5 pm. Evening campaign live from 4:45 pm with 55 percent, plus a 10 percent reserve that the account owner releases at 8:30 pm only if the evening campaign is pacing to exhaust before 10 pm. That reserve release is a two minute job and it is the single highest return two minutes in the week.

If your team cannot commit to three touch points a day, cut it to one. Hold 50 percent of the daily budget out of the campaign until 5 pm, then add it. That alone recovers most of the loss.

Availability gates the daypart, not the bid

Bidding harder into 9 pm is a waste if the SKU is not on the shelf. Quick commerce serves from the dark store nearest the customer, so availability is a store level fact, not a city level one. A brand can be at 92 percent availability on a city average and still be dark in the 30 stores that carry 40 percent of the city’s evening volume.

The rule we run: no daypart budget increase into a city where store level availability for the advertised SKU is below 85 percent, and no evening reserve release at all below 80 percent. Check availability at 4 pm, not at 9 am, because the morning and midday rush is what strips the shelf. If the number is short, the money moves to a city that can serve, not to a higher bid in a city that cannot. Paid clicks on an unavailable SKU do more damage than a lost impression, because the platform learns that your listing does not convert.

Pacing a Rs 3 lakh month

Take a monthly budget of Rs 3,00,000 across two platforms and four cities. Reserve 10 percent, Rs 30,000, for weekend evenings, which run 20 to 30 percent above weekday evenings and deserve their own line. That leaves Rs 2,70,000, or roughly Rs 9,000 a day.

Split it 20 percent morning, 15 percent midday, 55 percent evening, 10 percent late night on impulse SKUs only. That is Rs 1,800, Rs 1,350, Rs 4,950 and Rs 900 a day. Then weight by city on evening order share rather than on population. In most portfolios two cities carry 60 to 70 percent of quick commerce volume, so a four way even split is quietly funding two cities that will never scale.

Track cost per click and conversion separately per daypart. If evening cost per click is Rs 18 against midday at Rs 11, and evening converts at 9 percent against midday at 4 percent, evening is still cheaper per order by a wide margin. Keep pushing evening budget until that gap closes.

The twenty minute weekly review

Every Monday, four questions. First, which dayparts hit their budget cap before their window closed, which tells you where demand exceeded funding. Second, which SKUs took spend in hours where store availability was below 85 percent, which is pure waste to claw back. Third, what is cost per order by daypart by city, ranked, with the bottom quartile cut. Fourth, did weekend evening spend actually rise against weekday, or did the flat budget quietly flatten it again.

Fix those four and the same money buys 20 to 35 percent more orders without a single new keyword.