Skip to content
Case study

Zepto: can 10-minute delivery actually pay?

A unit-economics teardown of quick commerce's central question.

Breaks one order down to its contribution margin and identifies the two levers — basket size and orders-per-store-per-day — that decide whether the model survives.

Role
Product teardown — unit economics
Timeline
Strategy exercise
Year
2025
Domain
Quick commerce · Strategy
Unit economicsContribution-margin modelingLever analysis

The question under the hype

Forget growth. Does one order make money?

Quick commerce gets debated at the level of GMV and growth. The only question that matters underneath is the contribution margin of a single delivered order: take rate plus margin on the basket, minus the cost to pick, pack, and ride it to the door in ten minutes.

Model that one order honestly and the whole strategy falls out of it. Everything else is a lever on those few numbers.

The order, roughly (illustrative)

₹400–500
average basket — too small and nothing works
~20–25%
blended margin + take on the basket
₹40–60
last-mile cost per order, the line that hurts
1,000+
orders/store/day needed to dilute fixed cost

The two levers that decide it

What actually moves contribution margin

  • Basket size: last-mile cost is roughly fixed per order, so a larger basket spreads it. Every product nudge toward a bigger cart (bundles, thresholds, replenishment) is really a margin lever.
  • Orders per store per day: fixed dark-store cost only amortizes at density. This is why quick commerce is a winner-takes-density game fought neighborhood by neighborhood, not nationally.
  • Margin mix: private label and high-margin categories quietly fix the basket math better than any delivery optimization.

Tradeoffs

The strategic read

Chose

Win density block by block

over National coverage land-grab

The economics are local — a dark store either has the order density to amortize or it doesn't. Spreading thin maximizes the wrong number. Depth beats breadth here, structurally.

Chose

Treat basket size as a product surface

over Treating it as a marketing job

The product (bundling, replenishment, free-delivery thresholds) moves basket size more durably than promos, and it does it without burning margin on discounts.

Honest note

These are model numbers

The figures above are illustrative, chosen to be internally consistent, not pulled from Zepto's books. The point is the structure of the argument. [PLACEHOLDER: swap in real published figures if you cite this anywhere it matters.]

Next · case study

Rapido: the two-sided trust problem

Why bike-taxi growth is really a safety-perception problem.

Continue