Unit Economics: Does Each Sale Actually Make You Money?

A bucket factory can post fifty lakh rupees in sales and still be quietly going broke. If every bucket costs 110 rupees to make and sells for 100, the annual report looks busy and impressive while each bucket bleeds 10 rupees. The total hides the leak. A single bucket cannot.

That gap between what looks healthy and what actually is healthy has killed more exciting, fast-growing companies than any recession. And the only way to spot it is to stop staring at the big blurry total and pick up one bucket.

Why this matters

Here is the question that quietly decides whether a business lives or dies: when you make one sale, do you end up with more money than you started with?

It sounds obvious. It is not. Plenty of celebrated startups answered “no” without realizing it, and the faster they grew, the faster they failed. Their dashboards were green. Their revenue charts pointed up and to the right. The leak was hiding inside the average.

Unit economics is simply the math of one unit: one sale, one order, one customer. It is the act of picking up a single bucket and checking it for a hole. You do not need fancy software or an MBA. You need arithmetic you already know and the discipline to look.

The vocabulary, built one rung at a time

Each term below stacks on the one before it. The insight lives in how they connect, so do not skip a rung.

Revenue per unit

The price one customer pays for one purchase. Five hundred rupees for a subscription box this month. Just that one transaction, not the yearly total.

COGS, the cost of goods sold

The direct cost to make or deliver that one thing: raw materials, the ink and paper for a printed card, the cloud-server cost to serve one user.

It deliberately excludes fixed overhead like office rent and salaries, because those do not go up when you sell one more unit. Rent is the same whether you sell ten boxes or ten thousand.

Gross margin

The formula is (Revenue − COGS) ÷ Revenue. It answers a basic question: is the product itself worth making?

A 500-rupee product that costs 150 to make has a gross margin of 70 percent. That looks fantastic. Hold that thought, because it is also where most founders fool themselves.

Contribution margin, the honest number

Contribution margin is revenue minus all variable costs, not just COGS. That means shipping, the payment-gateway fee, returns, and the marketing you can attribute to winning that order.

It answers a sharper question: is it worth actually selling this, through this channel, at what it costs to win the customer? Contribution margin is always lower than gross margin, and the difference is rarely small.

Watch how a gorgeous 70 percent gross margin shrinks once you tell the truth:

Price the customer pays (Revenue)          ₹500
  − COGS (make and deliver it)            − ₹150   →  Gross margin = 70%
  − shipping                              − ₹30
  − payment-gateway fee (~2% + GST)       − ₹12
  − returns and refunds (averaged)        − ₹20
  − marketing attributed to this order    − ₹40
  ──────────────────────────────────────────────
  = Contribution margin (the honest one)    ₹248   →  ~49.6%

In harsher real-world cases, that same 70 percent gross margin can collapse to a 19.5 percent contribution margin. That fifty-point gap is exactly where a losing business gets mistaken for a winning one.

Adding the customer: CAC, churn, and LTV

One sale is just the start. Most real money comes from a customer who comes back. To measure that, you need three more terms.

CAC, customer acquisition cost

Total sales and marketing spend in a period, divided by the new customers you won in that period.

Crucially, this includes the salaries of the people doing sales and marketing, not just the ad bill. Founders who count only ad spend understate their CAC and feel richer than they are.

Churn rate

The percentage of customers, or revenue, you lose each period. If 5 percent leave every month, the average customer life is roughly 1 ÷ churn, which is 1 ÷ 0.05 = 20 months.

LTV, lifetime value

The total contribution margin one customer generates across their entire relationship with you. A simple and widely used formula, popularized by David Skok, is:

LTV = ARPU × gross margin ÷ churn rate

ARPU is just average revenue per user per period.

One trap snares almost everyone here: measuring LTV on revenue instead of margin. If a customer brings in 7,000 rupees of revenue but 4,000 of it is cost, that customer is worth 3,000 rupees of value, not 7,000. Always base LTV on contribution margin, or you will overstate every single customer.

The two ratios that deliver the verdict

Combine what a customer is worth (LTV) with what they cost to win (CAC), and you get the two numbers seasoned founders and investors actually judge a business on.

LTV to CAC, the magnitude check

The famous 3:1 benchmark comes from David Skok of Matrix Partners, drawn from mature public software companies. Read it like this:

• Below 1:1 means you lose money on every customer. Stop.
• Around 1 to 2:1 is fragile. One bad month can wipe you out.
• Around 3:1 is the healthy floor.
• 5:1 and above is strong, but can be a warning that you are underspending on growth and leaving money on the table.

CAC payback period, the speed check

The formula is CAC ÷ monthly contribution margin per customer. It measures how many months until you get your acquisition money back.

Two businesses can both sit at 3:1, yet a 6-month payback is dramatically healthier than an 18-month one. The first recycles its cash into the next customer long before the second has even recovered half. A healthy software target is a payback under 12 months.

The ratio tells you the size of the win. The payback tells you how long your cash is at risk. You need both. A beautiful ratio with a 20-month payback can still starve you of cash and kill you.

See it in one example

A subscription box:

• Price 500 a month, COGS 150, so contribution is 350 a month (70 percent).
• Monthly churn 5 percent, so lifetime is 1 ÷ 0.05 = 20 months.
• LTV = 350 × 20 = 7,000. CAC = 2,000.
• LTV to CAC = 7,000 ÷ 2,000 = 3.5:1. Healthy.
• Payback = 2,000 ÷ 350 = ~5.7 months. Healthy.

Now break it. Let churn double to 10 percent. Lifetime halves to 10 months, LTV drops to 3,500, and the ratio crashes to 1.75:1, fragile, with the same CAC and the same product.

Churn sits in the denominator, so a small worsening swings LTV violently. This is why fixing churn is usually your single highest-leverage move.

The most expensive lie in business: “we’ll make it up in volume”

If your per-unit economics are negative, growth does not save you. It accelerates the funeral.

The textbook case is MoviePass. From 2017 onward it charged about $9.95 a month for up to one movie a day, but it paid theaters $12 to $15 per ticket. It collected $9.95 and paid out $12 or more, a negative contribution margin on every active customer.

The result was perverse. The more people used the service, the faster it died. MoviePass rocketed past 3 million subscribers, which looked like a rocket ship, then burned $21.7 million in a single month, shut down in 2019, and its parent company filed for bankruptcy in 2020. Well over a billion dollars was destroyed.

Common misconceptions

“Amazon lost money for years, so losing money is fine.” This is the comparison people reach for, and it is backwards. Amazon’s per-transaction economics were positive. It deliberately spent on fixed investments like warehouses and software, ahead of scale. Losing money on fixed bets you will earn back is investing. Losing money on every transaction is subsidizing your own demise. The two are categorically different.

“Gross margin is my profit per sale.” It is not. Gross margin ignores shipping, fees, returns, and acquisition cost. Contribution margin is the per-sale number that tells the truth.

“A higher LTV to CAC ratio is always better.” Not necessarily. A 7:1 ratio can mean you are too timid with growth spend and a competitor is about to outrun you.

“The 3:1 rule applies to my brand-new startup.” It was never meant for pre-product-market-fit or seed-stage businesses. Early on, your CAC and churn numbers are tiny, noisy, and unreliable. Do not over-fit to them. Remember that LTV is a forecast built on an assumed churn rate, not a fact.

How to use this

Treat each number below as a lever you can pull. Work through them in order.

1. Calculate your true contribution margin first. Start with price, then subtract COGS, shipping, payment fees, returns, and attributed marketing. The number that remains is your honest profit per sale. If it is negative, nothing else matters yet.
2. Raise revenue or ARPU. Price increases, upsells, cross-sells, tiers, and bundles. This is usually the highest-leverage move, because most founders underprice out of fear.
3. Cut COGS. Negotiate better supplier terms, chase volume discounts, move to cheaper-to-serve infrastructure, automate. This widens both gross and contribution margin at once.
4. Lower CAC. Shift toward organic, referral, content, and word of mouth, and kill the channels that bleed. Analyze contribution margin per channel; one Google campaign can be profitable while another loses money.
5. Reduce churn. Better onboarding, a stickier product, annual plans. This is your highest LTV leverage. The dream is negative net churn, where existing customers expand their spending faster than others leave.
6. Shorten payback. Bill annually upfront and favor lower-CAC channels, so you recover cash faster and reinvest sooner.
7. If you are an Indian founder, bake in two costs everyone forgets. GST becomes a real cost layer once your turnover crosses the registration threshold (for FY 2025 to 26, roughly 40 lakh for goods and 20 lakh for services in normal-category states, lower in special-category states), so model your unit economics both below and above that line. Payment-gateway fees from providers like Razorpay or PayU run around 2 percent plus GST, which is about 12 rupees on a 500-rupee sale: small per order, but it belongs in contribution margin.

Conclusion

If you remember one thing, make it this: a single unit reveals the truth that the total hides. Before you celebrate a revenue chart, pick up one bucket and check it for a hole.

There is a beautiful problem waiting at the end of this road. If you ever reach negative net churn, where your existing customers grow faster than new ones leave, the simple LTV formula breaks because a customer’s lifetime becomes effectively infinite. At that point the arithmetic in this article runs out, and you graduate to a discounted-cash-flow model that accounts for expansion and a discount rate. That is a wonderful problem to have, and it is exactly where the most valuable businesses in the world quietly live.