CAC payback and LTV for Indian D2C: the real math

The 3:1 LTV to CAC rule was built for American SaaS. An Indian D2C brand with COD returns, marketplace fees and 20% repeat rates needs its own arithmetic.

A kirana shop in India with shelves of packaged goods stacked to the ceiling behind the counter
Photo: Renzilde, CC BY-SA 3.0, via Wikimedia Commons

For an Indian D2C brand, CAC payback is the number of months it takes for contribution margin from a customer's orders to repay what it cost to acquire them. Compute contribution per placed order after product cost, shipping, payment, marketplace fees and the expected loss from returned-to-origin (RTO) COD orders; then divide the CAC left after the first order by monthly repeat contribution. With COD RTO at 25 to 30% and repeat rates of 10 to 30% a year common in India, many brands that look profitable on ROAS never pay back their CAC at all.

Key takeaways
  • 60-65% of Indian e-commerce orders are cash on delivery, and 25-30% of those come back as RTO.
  • A ₹900 order can leave only about ₹328 of contribution, roughly 36% of what ROAS counts.
  • Payback months = CAC left after the first order ÷ monthly repeat contribution.
  • At 1.5 repeat orders a year, a ₹600 CAC pays back in about 6.6 months; at 0.3 it takes over two and a half years.
  • Quick commerce is a separate channel: the platform owns the buyer, so model its margin per unit, not as CAC.

Most D2C dashboards in India still show ROAS: revenue divided by ad spend. ROAS ignores what it costs to deliver an order, ignores orders that come back, and ignores whether the customer ever buys again.

LTV to CAC fixes some of this, but the textbook 3:1 rule came from subscription software with high gross margins and no returns. This post builds the numbers from the order up.

60-65%of Indian e-commerce orders paid by cash on delivery (Razorpay, citing ET Prime Research)
25-30%RTO rate on COD orders vs 2-3% on prepaid (Razorpay)
10-30%repeat purchase rates reported by Indian D2C founders (DSGCP, Meta and Viralmint survey, 2025)
270 million+online shoppers in India in 2024 (Bain)

What is wrong with the 3:1 LTV to CAC rule for Indian D2C?

The 3:1 rule fails in Indian D2C because it is usually computed on revenue, not contribution, and on a lifetime nobody can see. With these example inputs, an LTV of ₹6,000 in revenue at 60% gross margin, after shipping, payment fees and returns, may be ₹2,500 of contribution. Against a ₹1,500 CAC that is 1.7:1, not 4:1.

Same customer, two waysRevenue LTVContribution LTV
Lifetime value₹6,000₹2,500
CAC₹1,500₹1,500
Ratio4:1, "healthy"1.7:1, marginal

Three Indian realities make the gap bigger than in the US:

  • Cash on delivery. Razorpay puts COD at about 60 to 65% of Indian e-commerce orders, citing ET Prime Research, and higher in Tier 2 and Tier 3 cities.
  • Returns to origin. The same source puts RTO at 25 to 30% for COD orders against 2 to 3% for prepaid, and above 35 to 40% for COD fashion. It estimates the industry loses over ₹20,000 crore a year to RTO and failed deliveries.
  • Low repeat. A December 2025 survey of over 100 Indian D2C founders and operators by DSG Consumer Partners, Meta and Viralmint reported repeat purchase rates of 10 to 30%, and found 55% were under-investing in CRM and retention.

Add marketplace commissions and quick-commerce margins, and the "lifetime" in LTV becomes the most optimistic number in the deck.

Return-to-origin (RTO) rates in Indian e-commerce, by payment type
See the data as a table
Low end of range (%)High end of range (%)
COD orders2530
Prepaid orders23
COD fashion orders3540

Source: Razorpay: Cash on delivery and its challenges (updated Jun 2026) · Ranges as published; Razorpay does not cite a source for the RTO figures.

How do you calculate contribution margin per order in Indian D2C?

Contribution margin per order is what is left from the order value after every cost that scales with that order. It excludes fixed costs like salaries and rent, and it excludes acquisition cost, which is what you are testing it against. The costs that come out are:

  • product cost (COGS);
  • shipping both ways and packaging;
  • payment gateway or COD handling fees;
  • marketplace or platform commission;
  • the expected loss from orders that are shipped but never paid for (RTO).

A worked example on one ₹900 order

Here is a worked example for a ₹900 order at 60% gross margin, ₹120 of shipping, payment and packing per delivered order, 60% COD at 25% RTO and 40% prepaid at 3% RTO, with ₹150 lost on each RTO for forward and reverse freight.

From ₹900 order value to ₹328 contribution ₹900Order value -₹360Product cost -₹120Ship, pay, pack -₹92Expected RTO ₹328Contribution RTO: 16.2% of orders return (60% COD at 25%, 40% prepaid at 3%); each loses ₹420 margin plus ₹150 freight.
Illustrative numbers. Swap in your own in the calculator below; the RTO line is usually the one founders underestimate.

The RTO line works like this. The blended RTO rate is 0.60 × 25% + 0.40 × 3% = 16.2%. On each returned order you lose the ₹420 you would have earned plus ₹150 of freight, so the expected loss per placed order is 16.2% × ₹570, about ₹92.

Calculated in this post, with these example inputs, from the 60% COD share and the 25% and 3% RTO rates in the worked example, which sit inside the ranges Razorpay publishes (COD share 60 to 65%, RTO 25 to 30% on COD and 2 to 3% on prepaid).

Contribution per placed order is about ₹328, or 36% of order value. A ROAS dashboard showing "₹900 revenue" for that order overstates the money you can use to repay CAC by about 2.7 times.

Calculated in this post, with these example inputs, as the ₹900 order value divided by the ₹328 contribution worked out above.

The story

Gramiyaa, a cold-pressed oil brand from Chennai, began in 2018 as a family oil mill selling unbranded bottles through a handful of local shops. Its founders, Mohamed Yaseen, Naveenrajamaran RL and Sibi Manivannan, rebuilt it as a D2C brand in 2022, then spent about a year worrying over quick-commerce commissions before listing on Blinkit in 2024. Yaseen called it a year of anxiety. The channel lost money for roughly six months.

His own arithmetic, told to Inc42 in November 2026, is the contribution line in miniature. On a ₹400 sale the brand keeps about ₹300; the other ₹100 goes to platform margin and ads. As an unproven brand it paid about twice the commission of a national oil brand until volume earned it leverage, with operating margins of just 2-3% at the time. Quick commerce still became its biggest channel: about 35% of ₹29 crore in FY26 revenue, with customers reordering every 40 to 45 days.

Inc42: Why Gramiyaa had to think like a supermarket to win on quick commerce (Nov 2026)

How many months does it take to pay back CAC?

Payback months are the CAC left after the first order, divided by the contribution a customer produces per month from repeat orders. If the first order's contribution covers CAC, payback is immediate. If not, repeat behaviour decides everything.

Contribution-margin CAC payback for an Indian D2C brand

Use this to check what you just read. With the defaults, a ₹600 CAC leaves about ₹272 unpaid after the first order, and at 1.5 repeat orders a year it takes about 6.6 months to recover.

Calculated by the calculator above, with these example inputs (its default values).

Now drop repeat orders to 0.3 a year, closer to a 20% repeat rate, and payback stretches past two and a half years. Raise CAC to ₹1,200 with the same repeat and the brand is funding customers it will never earn back inside a typical funding runway.

Calculated by the same calculator, with these example inputs, changing only the repeat orders and then the CAC.

When does a customer pay back a ₹600 CAC? ₹0₹500₹1,000 CAC ₹600 1.5 repeats/yr: ₹820 by month 12 0.3 repeats/yr: ₹426 6.6 months ₹328 first order Month 0Month 12
The calculator's default case drawn as a cohort curve: first-order contribution of ₹328, then ₹41 a month from repeats. At 0.3 repeats a year the curve barely rises and stays under the CAC line.

Which costs kill D2C margins on the way to ₹100 crore?

The costs that kill margins as a D2C brand scales are mostly variable costs that grow faster than revenue: acquisition, returns, channel fees and discounts. The table sets out where each one hits the unit economics.

Cost lineWhere it hitsWhy it grows with scaleWhat to measure
Meta and Google CACPayback monthsNew audiences are less ready to buy; over 70% of surveyed founders rely on Meta as the primary channel (DSGCP 2025)First-order contribution minus CAC, by cohort
COD RTOContribution per placed orderAbout three in five new shoppers since 2020 came from Tier 3 and smaller cities (Bain), where COD share is higher (Razorpay)RTO by pin code, payment type and creative
Marketplace commissionContribution per orderMarketplaces bring volume at a take rate on every order, including repeatsContribution by channel, not blended
Quick-commerce margins and adsContribution per orderPlatform margin plus on-platform ads to stay visibleNet realisation per unit after ads
Discounts to drive repeatRepeat contributionRetention offers train customers to wait for a codeRepeat rate with and without discount
Creative productionCAC62% of surveyed founders report creative fatigue (DSGCP 2025)Cost per winning creative

The pattern behind the ₹100 crore wall is covered in why Indian D2C brands stall before ₹100 crore. The same survey found about 60 to 65% of brands stuck in the ₹1 to 50 crore band.

A street shop decorated for Diwali with lanterns and festive goods on display
Festive peaks bring cheap volume and expensive customers. Cohort payback shows whether Diwali buyers ever come back.Photo: आर्या जोशी, CC BY-SA 4.0, via Wikimedia Commons

How does quick commerce change D2C unit economics?

Quick commerce trades acquisition cost for platform margin: the platform brings the buyer, and the brand pays for it on every unit, first order or tenth. Bain's How India Shops Online 2025 report says quick commerce made up over two-thirds of e-grocery orders and about a tenth of e-retail spend in 2024, and expects it to grow over 40% a year to 2030.

What the Gramiyaa numbers show

Inc42's November 2026 profile of Gramiyaa, told in the story above, gives rare numbers. Quick commerce was about 35% of its ₹29 crore FY26 revenue. On a ₹400 product, the brand keeps about ₹300 and roughly ₹100 goes to platform margin and ads; its commission was around twice what national brands pay until volume gave it leverage.

Customers reorder every 40 to 45 days. The same month, Inc42 reported beauty brands such as Moxie Beauty getting over 30% of sales from quick commerce.

How to model it

QuestionYour own site or WhatsApp listQuick commerce
Who owns the customer?YouThe platform
How you pay for the buyerCAC, oncePlatform margin and on-platform ads, every unit
Does repeat lower future acquisition cost?YesNo
What to trackCAC payback by cohortContribution per unit after margin and ads

For CAC math, this means two things. On quick commerce there is little measurable "CAC" in the D2C sense, because the platform owns the customer; you track contribution per unit after platform margin and on-platform ads.

And repeat on quick commerce does not lower your future acquisition cost the way an owned customer on your website or WhatsApp list does. Model it as a separate channel with its own contribution, not as part of your D2C LTV.

How do you estimate repeat rate and LTV from your own data?

Estimate repeat behaviour from cohorts, not averages: take every customer who first bought in a given month, and track how many bought again in month 1, 2, 3 and onward. A probabilistic model can then project future purchases.

Two classic papers from Peter Fader and Bruce Hardie show how, using only data most Shopify stores already have: recency, frequency and how long each customer has been with you.

"Counting Your Customers" the Easy Way: An Alternative to the Pareto/NBD ModelPeter S. Fader, Bruce G.S. Hardie, Ka Lok Lee · Marketing Science · 2005

The authors propose the BG/NBD model, which predicts each customer's future purchases and the probability they are still "alive" from three numbers: how many repeat purchases they made, when the last one was, and how long they have been a customer. They test it on CDNOW, an online music store: a cohort of 23,570 first-time buyers from early 1997 who bought nearly 163,000 more CDs over 78 weeks. Calibrating on a 2,357-customer sample over 39 weeks, the model forecasts the next 39 weeks about as well as the harder Pareto/NBD model, and it can be fitted in a spreadsheet. The limit: it assumes non-contractual buying with no seasonality, so a Diwali-heavy brand needs adjustments.

Read the paper (free)
RFM and CLV: Using Iso-value Curves for Customer Base AnalysisPeter S. Fader, Bruce G.S. Hardie, Ka Lok Lee · Journal of Marketing Research · 2005

This paper links recency, frequency and monetary value (RFM) directly to customer lifetime value, using a Pareto/NBD model for purchase counts and a separate model for spend per order, again on the CDNOW cohort. The average customer's discounted lifetime value was about $47, but the top RFM cell averaged $435 and held nearly 38% of the cohort's future value. A counter-intuitive finding: customers who bought often but not recently were worth less than lower-frequency customers who bought recently, because their silence suggests they have left. The limit: 1990s music retail, so the numbers do not transfer, but the method does.

Read the paper (free)
The top RFM cell held nearly 38% of the cohort's future value. LTV is never spread evenly.Fader, Hardie and Lee, Journal of Marketing Research (2005)

The practical takeaway for an Indian D2C brand: LTV is concentrated, so payback should be modelled by acquisition cohort and by channel, not as one blended number. If your Meta cohorts from festive sales have half the 90-day repeat of your organic cohorts, they have a different CAC ceiling.

Feed those cohorts into measurement properly, as covered in how to build a measurement stack after cookies, and use a marketing mix model to check whether a channel adds customers or only claims them.

Front of a small kirana store in India with goods displayed at the entrance
The neighbourhood kirana has always run on repeat customers. D2C brands have to earn the same habit with ads and CRM.Photo: नमन भगत, CC BY-SA 4.0, via Wikimedia Commons

What CAC payback should an Indian D2C brand target?

There is no universal number, but the target should be shorter than the time your cash can wait and shorter than the time a cohort reliably keeps buying. If 90-day repeat drops off sharply after month three, a 12-month payback is a bet on customers you are already losing.

A sensible way to set it:

  1. Measure first-order contribution by channel. Use the calculator with your real RTO, fees and shipping. If it is negative, fix it before scaling ads.
  2. Build monthly cohort curves. Cumulative contribution per acquired customer, month by month, for each channel and season.
  3. Find where curves flatten. That month is the practical edge of your "lifetime".
  4. Set the CAC ceiling below the flattening point's cumulative contribution, with a buffer for cost of capital and settlement delays.
  5. Attack the cheapest lever first. Cutting COD RTO, covered in how to reduce COD RTO, often improves payback faster than lowering CAC.

Brands selling food and other consumables, where reorder cycles are short, can afford longer payback than gifting or furniture brands; see how a D2C food brand grows online sales in India. And because ad cost per sale depends on the creative system as much as on targeting, read the creative testing framework for Meta alongside this.

I build these models for brands that want a growth plan finance will sign off. If that is useful, the hire page says how I work.

Questions people ask

What is a good LTV to CAC ratio for D2C in India?

Compute it on contribution margin over a horizon you can observe, such as 12 months, not on revenue over an imagined lifetime. On that basis, a ratio comfortably above 1 within 12 months, with payback well inside your cash runway, is a healthier test than a revenue-based 3:1.

Should I count COD orders in ROAS?

Count them only once delivered and paid, or apply your RTO rate to them. Platform-reported ROAS usually counts the order at placement, which inflates it by roughly the RTO rate on COD-heavy traffic.

How do I grow online sales for a D2C food brand without burning cash?

Food brands win on repeat, so focus on first-order contribution and 60 to 90 day reorder rates by channel. Quick commerce can deliver volume, but model its platform margin and ads as a per-unit cost rather than as acquisition.

Is CAC payback or LTV more useful for a founder?

Payback is more useful day to day, because it is observable and tied to cash. LTV projections are useful for valuing cohorts, but they depend on assumptions about the future that early-stage brands rarely have data to support.

Sources

  1. Razorpay: Cash on delivery and its challenges (updated Jun 2026)
  2. Entrepreneur India: Why most Indian D2C brands fail to cross INR 100 crore (Dec 2025)
  3. Bain & Company: How India Shops Online 2025 (Mar 2025)
  4. Inc42: Why Gramiyaa had to think like a supermarket to win on quick commerce (Nov 2026)
  5. Inc42: Beauty brands go quick commerce-first (Nov 2026)
  6. Inc42: Gramiyaa founders' account of listing on Blinkit (source for the story) (Nov 2026)
  7. Fader, Hardie and Lee: "Counting Your Customers" the Easy Way, Marketing Science (2005)
  8. Fader, Hardie and Lee: RFM and CLV, Using Iso-value Curves for Customer Base Analysis, Journal of Marketing Research (2005)
CACLTVD2Ccontribution marginRTOcash on deliveryquick commerceunit economics
Sapta

Sapta (Saptarshi Sarkar) is a mechanical engineer turned marketer in Bengaluru and a solo touring rider. He writes the seven Saptaverse blogs. Work with him or read more of Marketblog.

Keep reading in Marketblog

From the other blogs