पाठशाला Pathshala · हिसाब Hisāb, Unit economics · Lesson 05 · Start

D2C unit economics: the order that has to make money

A brand does not sell products. It sells orders, one at a time, and each one either makes money or does not. The per-order P&L with returns, logistics, discounts and marketplace fees, and the lines where the margin disappears.

Pathshala, The Founder Library · 11 October 2026 · 8 min read

A direct-to-consumer brand does not sell products. It sells orders, one at a time, each with its own courier, its own box, its own chance of coming back and its own share of the advertising that produced it. Each order either makes money or does not, and a brand is nothing more than the sum of those answers. The founder who knows the answer per order can scale. The one who knows only the gross margin is about to learn why the bank balance disagrees with the spreadsheet.

This lesson builds the per-order profit and loss line by line, gives the Indian numbers for the two lines that surprise founders most, returns and marketplace fees, shows what the listed D2C companies’ accounts say about the distance between gross margin and profit, and puts the whole thing in a figure that switches between a brand’s own site and a marketplace. It ends with the three levers that move the result and a monthly ritual.

The per-order P&L, line by line

Begin with the order value and strip GST, because the tax the customer pays is collected for the government and is not revenue; a brand that computes margin on the inclusive price has overstated it before the first cost. Take off the discount, since nearly every D2C order carries one. Then take the lines in the order they happen. Cost of goods on the units the customer keeps, since a returned unit goes back to stock. Payment: on the brand’s own site a gateway fee of about two per cent plus GST on prepaid orders, which is Razorpay’s stated standard rate, and a cash-collection charge from the courier on cash-on-delivery orders. Marketplace fees if the order was placed on one. Courier, forward on every order and back again on every return. Packaging. Revenue lost on returns, because the refund goes out and the sale is undone. And finally marketing per order, the advertising it took to produce this order, which is the line most brands keep below contribution and should keep in sight.

A skincare brand in Hyderabad sells a ₹1,200 order ex-GST at a fifteen per cent discount, so net price is ₹1,020. Cost of goods is thirty per cent of net price. Eight per cent of orders come back, forty per cent are paid in cash on delivery at a ₹45 collection fee, the courier charges ₹75 a leg and the box costs ₹30. Revenue kept per order is ₹938. Cost of goods on the kept units is ₹282, gateway and COD fees are ₹32, the courier is ₹81 with the return legs, packaging ₹30. Contribution before marketing is ₹513, a healthy 55 per cent of revenue kept. The brand spends ₹350 of advertising to win each order. What is left is ₹163, seventeen per cent of revenue kept, from a product whose gross margin on the slide was seventy per cent.

Returns: the Indian line

Returns are where Indian D2C economics differ most from the American playbooks founders read. Unicommerce’s India Ecommerce Index 2023 measured returns at 10.4 per cent of all orders in FY23, up from 9.8 per cent the year before. The average hides two splits that matter more than the average. By channel, marketplace orders were returned at 26.3 per cent against 6.2 per cent on brand websites. By payment, cash-on-delivery orders were returned at 20.9 per cent against 5.8 per cent for prepaid. A brand that sells on marketplaces to cash-paying customers is running a business in which one order in four does not stick, and every one of those orders cost two courier legs.

The arithmetic of a return is worse than it looks. The forward courier is spent. The reverse courier is spent. The product may come back damaged or unsaleable, which in cosmetics and innerwear means it is written off. The gateway fee is often not refunded. And the customer who returned has still been counted in the advertising cost that won the order. A ₹75 courier each way on a ₹1,020 order is fifteen per cent of net price gone for nothing. Return rate is therefore the single most powerful slider in the figure below, and the brand that measures it by product, by pin code and by payment mode will find that a small number of combinations produce most of the losses.

Marketplace fees: what the platform charges

Selling on a marketplace replaces the gateway and the brand’s own traffic cost with the platform’s fees, and those are several. Amazon.in’s fee page lists four kinds. A referral fee that starts at zero and varies by category, with examples from five per cent on mobile phones through seven per cent on baby apparel above ₹1,000 to thirty per cent on drones. A closing fee that starts at ₹2 and rises with price and fulfilment channel, with examples up to ₹75 for some categories above ₹1,000. A weight-handling fee that starts at ₹37 per item and rises with weight and distance. And fees for optional services such as pick-and-pack and storage if the platform fulfils. GST at eighteen per cent applies to all of them. Since March 2025 referral fees on products priced below ₹300 have been zero, which has changed the arithmetic for low-ticket categories.

Add the lines together for a ₹1,000 beauty or apparel order and the platform’s total share commonly lands between fifteen and twenty-five per cent before advertising on the platform, which is a second cost that sellers in competitive categories find they cannot avoid. The figure uses eighteen per cent as the marketplace default. Check your own category on the fee page, because the referral rate alone can move the per-order answer from positive to negative.

What the listed brands’ accounts say

The distance between gross margin and profit is not a seed-stage problem that scale solves. Honasa Consumer, the Mamaearth parent and the largest listed Indian D2C company, reported a gross margin of 70.7 per cent and an EBITDA margin of 5.1 per cent for the quarter ended March 2025. Sixty-five points of margin went between the product and the operating line. The largest single destination was advertising: the company spent ₹744 crore on advertising and marketing in FY25, 36 per cent of revenue. FSN E-Commerce, which runs Nykaa, spent ₹958 crore on marketing and advertising on ₹7,950 crore of revenue in the same year, about twelve per cent, and it owns the storefront most of its brands sell through.

Read those two numbers as a founder, not as an analyst. A brand at seventy per cent gross margin that spends a third of revenue on advertising and a fifth on logistics and platform fees arrives near zero before salaries. That is the per-order P&L of this lesson at the scale of a listed company, and it is why the order has to make money before marketing, with a margin wide enough to pay for the marketing and leave something behind.

The figure opens on the Hyderabad brand selling on its own site. Switch the channel to marketplace. Returns rise to a quarter of orders, the platform takes eighteen per cent of what is kept, and the same ₹350 of advertising now produces a loss on every order. Switch back and move the cash-on-delivery share to a hundred per cent, then the return rate to the twenty-one per cent that COD orders actually show, and watch what is left after marketing fall from ₹163 to under ₹50 without changing channel. The product never changed. The order did.

Gross margin is what the brand would make if nothing went wrong between the warehouse and the doorstep. Contribution is what it makes given that something always does.

The three levers

Order value. Courier, packaging and the COD charge are fixed per order, so a larger order carries them more easily. A ₹75 courier is seven per cent of a ₹1,020 order and four per cent of a ₹1,800 one. Bundles, a free-shipping threshold set a little above the current average order and a second product that the first one makes necessary are the usual ways to move it. Raising the order value by half does more for the per-order answer than any discount ever has.

Prepaid share. Every COD order costs a collection fee and returns at three to four times the prepaid rate. A prepaid discount of two per cent is cheaper than the COD charge and the return it prevents, and brands that have pushed prepaid share from forty to seventy per cent have watched their return line fall without touching the product. Address and phone verification before dispatch on COD orders does the same for returns to origin.

Marketing per order. This is the line founders control least and spend most on. Compute it honestly as total advertising divided by orders won through advertising, not blended with repeat orders that would have arrived anyway; the lesson on [CAC, LTV and payback](/library/cac-ltv-and-payback-the-three-numbers) has the method. Then note that a repeat order carries no marketing line at all. A brand whose customers order three times a year has three orders to carry one marketing cost, and the only route to the margins the slide promised runs through repeat purchase.

A monthly ritual, in thirty minutes

On the fifth working day of each month, once the courier and marketplace settlements are in, build the per-order P&L for the previous month four times: own site prepaid, own site COD, marketplace, and repeat orders from existing customers. Use actual courier invoices, actual return counts and the actual settlement statements from each platform, because the fee you were quoted and the fee you were charged differ by the closing fees and the GST that quotes leave out. Put the four numbers beside last month’s four.

Then apply one rule. Any segment in which the order loses money after marketing is a segment to shrink, fix or price differently before the next advertising rupee goes into it, however fast it is growing. Any segment in which the order makes money after marketing is one in which the question becomes how much faster you can afford to grow, and that is the question the lesson on [contribution margin](/library/contribution-margin-first-number-to-know) and the one on runway were written to answer.

Nothing here is legal, tax or investment advice. Fee schedules and courier rates change; the ones cited were checked in October 2026 and your own settlement statements are the final word.

Sources

  1. Unicommerce, India Ecommerce Index 2023 — Returns at 10.4 per cent of orders in FY23; marketplaces 26.3 per cent against brand websites 6.2 per cent; COD 20.9 per cent against prepaid 5.8 per cent.
  2. Amazon.in Seller Central, Fees and pricing — Referral, closing, weight-handling and other fees with worked examples; GST at 18 per cent on all fees. Checked October 2026.
  3. Amazon India, Zero referral fees on over 1.2 crore products priced below ₹300, March 2025
  4. Business Standard, Honasa Consumer shares fly 14% on Q4 results, May 2025 — Gross margin of 70.7 per cent and EBITDA margin of 5.1 per cent for the quarter ended March 2025.
  5. Storyboard18, Mamaearth parent ups marketing spend to Rs 744 crore in FY25, September 2025 — Advertising and marketing at 36 per cent of FY25 revenue of ₹2,067 crore.
  6. exchange4media, Nykaa ad spends up 33% to Rs 958 cr in FY25, May 2025 — FSN E-Commerce marketing and advertising of ₹958 crore on revenue from operations of ₹7,950 crore.
  7. Razorpay, Payment Gateway Pricing and Fees Explained, February 2026 — Standard domestic rate of 2 per cent plus 18 per cent GST on the fee.