29 percent of order value goes to discounts at one quick-commerce network, while 1,240 dark stores run at 1,192 orders a day each. Those two figures describe the same business, and the second sets the first: what a store can give away is what its throughput leaves after the last mile is paid for. Swiggy's September-quarter disclosure put Instamart's net order value at 71 percent of gross order value, down from 74 percent in the prior quarter, with contribution margin at minus 1.0 percent of gross order value and a network of 1,240 active dark stores holding 5.21 million square feet. For brands and retail teams that fund promotions through delivery platforms, the useful unit is no longer the city or the campaign but the delivery bay, as of November 12, 2026.
What a bay can give away
The company's own definitions make the arithmetic traceable. Gross order value is counted at maximum retail price before any discount and includes delivery charges, packaging, user fees and taxes, excluding tips; net order value subtracts every discount, whoever funded it — platform, brand or seller. The distance between the two lines is, by construction, the discount itself: 26 percent of order value a quarter earlier, 29 percent now. Everything a brand negotiates about promotion funding sits inside that gap, and the gap is reported as a network average rather than a store figure.
The 29 percent nobody prices
Swiggy's shareholder letter notes that discounts on non-grocery items are typically much higher than on grocery, which turns the ratio into a mix indicator as much as a price indicator. A bay that adds personal care and small electronics to its assortment will look more promotional without a single change to grocery pricing, and the reverse is equally true: a store trimming non-grocery lines can appear to have tightened discount discipline while doing nothing of the kind. That is the first argument for measuring discount depth per bay, because the number moves with the assortment decision and not only with the campaign calendar.
Where the discount leaks out first
The cost side of the same equation is set by geography, and parcel pricing shows the spread. A residential delivery in the United States carries a surcharge of 6.45 dollars at FedEx and 6.50 dollars at UPS on top of the transport charge; fuel is priced as a weekly percentage of that charge, 29.25 percent at FedEx Ground and 30.25 percent at UPS Ground in the week of November 2, 2026; delivery area surcharges apply again for addresses outside dense zones, and one UPS product raised that fee 61.8 percent, from 3.80 to 6.15 dollars.
| Cost line | What it adds | Where it bites |
|---|---|---|
| Residential surcharge | 6.45 to 6.50 dollars per parcel | Every doorstop order |
| Fuel surcharge | 29 to 30 percent of the transport charge | Long routes, small baskets |
| Delivery area surcharge | Up 61.8 percent on one UPS product | Low-density catchments |
| Dimensional weight | Billed on volume, not on weight | Bulky, light products |
Stacked together, surcharges of this kind account for roughly a third of an average package cost, and the effective increase for direct-to-consumer shippers landed between 10 and 18 percent in 2026 even where headline rate rises were smaller. The lesson travels beyond parcel networks: two bays that promise the same twenty-minute window can carry very different delivery costs, so a discount budget that is affordable at the network average is unaffordable at the edge of the map.
Cost per stop decides the ceiling
Two bays, two budgets
Zone pricing makes the gap explicit. A 20-pound parcel costs 24.20 dollars to a nearby zone and 72.15 dollars to the farthest one on the same postal service, a threefold spread that no promotion can absorb. Over the same period shipping cost as a share of revenue rose about 90 basis points year on year for merchants turning over 5 to 50 million dollars, and the practical free-shipping threshold now sits at 1.3 to 1.5 times average order value. Set beside the quick-commerce numbers, the ceiling on discount depth is throughput minus area cost, and only one of those two appears in a marketing plan.
For a brand that funds promotions through a delivery platform, the consequence is a change in the unit of analysis. A city-level discount target hides the bay where contribution is negative; a bay-level target makes the trade-off visible, because the same unit of currency buys fewer orders where the last mile costs more. It also changes what a brand asks for at the next negotiation: not a national slot, but a store cluster with a defined delivery promise and a measured fill rate attached to it.
A short list before the next campaign
Three checks make the next funding decision auditable without waiting for a new disclosure cycle. Each is a question a category team can put to a platform or marketplace partner this week, and each answer should come back as a number rather than an opinion.
- Ask for net order value and gross order value by store cluster, not by city, and recompute the discount line before approving any promotion funding.
- Match each discount proposal to the delivery cost of the cluster it targets, surcharges and fuel included, so the contribution test uses the real last-mile figure.
- Track assortment mix next to discount depth, because adding non-grocery lines lifts the ratio without a single change in grocery shelf prices.
None of this makes discounts smaller on its own. It makes them locatable, which is the difference between a budget defended in a meeting and one discovered in a quarterly filing. The next disclosure will show whether contribution margin recovers at the same pace as order value; until it does, the bay remains the only unit where the two numbers can be compared honestly.
FAQ
Why compare net order value with gross order value instead of revenue?
Because gross order value is counted at maximum retail price before discounts, while net order value subtracts every discount funded by the platform, a brand or a seller. The ratio therefore isolates the discount line, which revenue figures bury inside net sales.
Is 1,192 orders per store per day a capacity ceiling?
It is a utilisation measure computed on active store-days, not a limit. It counts orders against the days a store was active, so a store that opened or closed inside the quarter barely moves the average, which is why the number flatters new networks.
Do United States parcel surcharges apply to Indian quick commerce?
Not directly. They are used here as evidence of how area-specific the last mile is: the same parcel costs three times more to a distant zone, and that spread is what a per-bay discount budget has to survive.
What can a brand control when the platform sets the discount?
Three things: which categories it funds, how deep the funding goes per cluster, and whether the assortment change that shifts the ratio is priced in. The rest is reported after the quarter closes.
