As of November 8, 2026, the clearest signal from India's festive peak is not that delivery apps ran out of riders. It is that availability failures were billed to shoppers as a pricing event. On November 6, Blinkit, Swiggy Instamart and Zepto all imposed temporary service limits, longer delivery windows or surge fees during the Dhanteras rush, and every one of those tools is easier to see on an app screen than an item that was never listed in that store.
What the Dhanteras surge actually rationed
Surge pricing and minimum order values protect unit economics during a spike, and platforms are entitled to use them. The problem for a brand is what those signals hide. When the fee goes up, the shopper reads the friction as a price event and either pays or waits. Neither response produces a stockout complaint, so the brand's dashboard records a demand dip rather than an availability failure in the pincode it paid to serve.
The scale behind the peak is not in dispute. Quick commerce is projected to account for close to 18 percent of festive e-commerce in 2026, against 11 percent a year earlier, and the festive basket has moved well past groceries into fashion, beauty, gifting and precious metals. A network built to pick twenty grocery lines per order in a few minutes is now asked to pick a gift with size, variant and packaging constraints from the same shelf.
Surge fees mask an availability problem
Three levers were used in the same week and each one changes a different number. Temporary service suspension removes the store from the app entirely; an extended delivery window lowers the promise without touching the assortment; a surge fee raises the price of the same basket. Only the first is visible as an availability event, and even then it tends to be logged as a platform issue rather than a brand issue.
| Platform action | What the shopper sees | What the brand's report shows |
|---|---|---|
| Temporary service suspension | the store is not selectable | zero orders, no out of stock flag |
| Extended delivery window | a longer promise at the same price | lower conversion, flat availability |
| Surge fee or minimum order value | a higher basket or a fee line | lower conversion, higher average basket |
| Item never listed in that store | the item simply does not appear | nothing at all |
The last row is the one worth designing around. A quick commerce interface hides unavailability by omission: if a store does not carry the item, it does not appear for that pincode, and the shopper concludes the platform does not sell it. There is no out of stock label to scrape, no substitution prompt and no complaint to route, which is why listing coverage has to be measured against the intended assortment rather than read off a platform report.
The festive basket widened faster than the shelf
Category data from the same window shows how far the basket moved. On Swiggy Instamart, gold coin orders rose 23 times in the week before Dhanteras, mithai orders grew more than 55 percent, fresh flowers 42 percent and LED diyas 88 percent, with protein chips up 239 percent. Flipkart Minutes recorded a threefold order increase between October 30 and November 5, with diyas up twelve times and kitchen appliances up ten times.
What that means for a store with a four minute pick window
Every one of those categories carries attributes a grocery pick path was never designed for. A coin has a denomination, a mithai box has a shelf life, a diya has a fragile shape and a gift hamper has a variant. When the pick path cannot resolve the attribute, the item is dropped from the order or the order is cancelled, and both outcomes look like a demand problem from the outside.
Speed is now a conversion variable, not a delivery promise
Agency data from the same season makes the cost of that ambiguity explicit. HiveMinds reported that on a leading marketplace, conversion for a category fell from 4 percent to 2 percent when delivery timelines moved from next day to four to seven days, which is the first hard number that treats speed as a conversion input rather than a fulfilment metric. Click costs on Flipkart ran up to 2.5 times normal during the sale, and generic searches lost about half their efficiency while branded searches held.
The same review found brands matching competitors' prices selectively instead of cutting across the board, and quick commerce offtake for a grocery and wellness brand growing 10 percent through October while dry fruits, a classic gift, slipped to marketplaces. Read together, the pattern is that the shelf fight and the bid fight produced different winners, and neither is visible in a blended growth number.
A pre-peak checklist for brands selling on delivery apps
The checklist below is designed for the two weeks before a peak, because that is when listing rules and assortment decisions are still open. It assumes the brand already receives some form of platform reporting and does not add a new data source; it only adds a comparison the brand currently does not run.
Who checks what, and which evidence settles it
FAQ
Did the platforms actually run out of capacity on Dhanteras?
They ran into delivery constraints and answered with limits, longer windows and surge fees. That is a capacity signal, but it is not the same as an availability signal, and the two get merged in most brand reports.
Why is a stockout hard to see on a quick commerce app?
Because the interface hides unavailability by omission. An item a store does not carry never appears for that pincode, so there is no out of stock label, no substitution prompt and no complaint to track.
Is 18 percent of festive e-commerce a reliable share figure?
It is a projection for 2026 against 11 percent in 2025. Treat it as a direction rather than a precise number, and check whether the source measures order value or order count.
What is the single cheapest check a brand can add?
A weekly diff between the intended SKU list and the visible assortment for the pincodes that matter, run on the same weekday each week. It costs one analyst hour and surfaces failures no platform report contains.
