Unit economics for marketers: growth that pays

What is the unit in unit economics?

The unit is the specific piece of business whose economics you are evaluating. It might be an order for an online shop, a customer for a repeat-purchase business or a contract for a service provider.

Choose the unit before building the spreadsheet. Dividing monthly advertising spend by orders gives a different answer from dividing it by newly acquired customers. A returning customer can place several orders without being newly acquired several times.

Define the period too. A customer's first purchase, observed contribution over a fixed window and projected lifetime value are distinct measures. Place them beside one another if useful, but label what is observed and what is assumed.

This article concerns the commercial logic behind marketing choices. The broader marketing finance guide covers the vocabulary, while the D2C acquisition-cost article focuses more narrowly on acquisition.

Why does revenue growth need a unit-economics check?

Revenue growth can consume cash and reduce profit when the costs of acquiring and serving demand rise faster than the contribution available. A successful-looking campaign can amplify an uneconomic offer.

The US Small Business Administration's break-even guidance, checked in October 2026, calculates break-even using fixed costs and the difference between price and variable cost. The key lesson for a marketer is that revenue alone cannot establish whether additional volume helps cover the business's costs.

The IFRS Foundation's IFRS 15 overview distinguishes contractual performance from simply receiving money. Ask finance how revenue should be represented in the model; a prepaid service may still create delivery obligations.

Google Analytics' ecommerce documentation distinguishes event-level and item-level measures. This is another reason to reconcile reporting definitions. A count of purchase events is not automatically a count of products, customers or financially settled orders.

How do you calculate useful unit economics?

Begin with net revenue, subtract clearly defined variable costs and then examine acquisition and fixed costs separately. Keep the model transparent enough that a colleague can reproduce it.

1. Establish net revenue

Start with the actual commercial amount relevant to the unit. Account for discounts, refunds and cancellations, and agree the treatment of taxes with finance. Do not compare a tax-inclusive platform value with a tax-exclusive cost model without explaining the difference.

Reconcile a sample of records across the store or CRM, payment system and financial records. Check duplicated transactions and the timing of returns. A reliable model often begins with a few carefully inspected records rather than an elaborate dashboard.

2. Identify costs that change with the unit

For an order, relevant costs may include the product, packaging, payment processing, delivery, expected returns and variable support. For a service contract, include the delivery effort that increases when the contract is added.

Do not assume every salary is fixed in every decision. Existing capacity may absorb a small increase, while a larger increase requires hiring. Document where that threshold changes the economics instead of projecting the same cost indefinitely.

3. Calculate contribution before acquisition

Consider an illustrative order with ₹1,200 in net revenue. Assume ₹500 product cost, ₹80 packaging, ₹120 delivery and ₹100 for payment, expected returns and variable support combined. These are teaching assumptions, not market benchmarks. Contribution before acquisition is ₹400.

If acquisition allocated to that new customer is ₹300, the first order leaves ₹100 before fixed overhead. That does not establish net profit. It shows the amount available under the defined boundary.

Now assume a promotion reduces net revenue by ₹150 without changing the other costs. Contribution after acquisition becomes negative ₹50. A higher conversion rate might still justify the promotion, but only if the additional volume and later economics compensate. The original margin cannot remain in the forecast after the price changes.

4. Build customer cohorts

Group customers by acquisition period and relevant offer or source. Observe their subsequent net revenue, contribution and service costs over comparable windows. Keep customers who never return in the cohort rather than analysing only repeat buyers.

A newer cohort has had less time to purchase again. Compare equivalent ages or label the difference. Avoid assuming that the most recent acquisition channel will eventually behave like older customers acquired under different conditions.

5. Estimate payback with timing

Payback asks when accumulated contribution recovers the acquisition investment. Show the cash implications separately when collection and delivery timing differ.

A business may report attractive projected lifetime value while struggling to fund the period before recovery. Consider delayed payments, inventory purchases and refunds. A model should help the team understand the timing of pressure, not only the eventual total.

6. Stress-test the decision

Change the assumptions most likely to break the plan: return rate, contribution, repeat frequency, acquisition cost or delivery capacity. Show what decision changes when an assumption moves.

Do not automatically use a universal lifetime-value-to-acquisition-cost target. A ratio can conceal different time horizons, risks and cost definitions. Establish an acceptable range with finance based on the business's cash, uncertainty and objectives.

Where do marketers misuse unit economics?

They misuse it when they combine incompatible populations or omit costs that become visible outside the marketing report. Average order revenue divided by customer acquisition cost is not a complete measure of customer value.

In Indian commerce, include the actual implications of payment methods, failed delivery and returns where they apply. Do not assume that a recorded order became a completed, retained sale. Use the business's own evidence rather than a generic industry rate.

B2B services have different complications. A contract can look valuable while requiring substantial onboarding or custom work. Include delivery commitments and collection timing, and distinguish one-time implementation revenue from recurring revenue.

Channel comparisons also need caution. A referral source may bring fewer but more suitable customers; a broad campaign may create future demand outside the immediate reporting window. Unit economics is an essential decision input, but it should not be used to pretend every causal question has been solved.

When should the model change?

Update the model when price, cost, customer mix or observed behaviour changes materially. Review it before increasing spend into a new audience or offer.

In week one, agree definitions with finance and inspect sample records. Next, calculate a simple order or contract view. During the first 30 days, establish an observed cohort report and document which assumptions still lack enough history.

Allow returns and delayed costs to mature before treating a cohort's contribution as final. Keep provisional reporting useful by labelling it, rather than waiting indefinitely or presenting early values as settled facts.

Who should own unit economics?

Finance should own the accounting definitions and cost treatment, while marketing owns how the model informs acquisition and offer decisions. Operations and sales contribute the costs and conditions of delivery.

Analytics should reconcile identifiers and reporting windows. Leadership should decide which risks the business can fund. No department should be able to improve its apparent performance merely by moving a cost outside its preferred definition.

Store the definitions beside the calculation. A model that only one analyst understands will drift when staff, tools or offers change.

My view

I would prefer a modest, observed contribution estimate to a spectacular lifetime-value number built on untested repeat behaviour. The trade-off is that conservative assumptions may make an early growth proposal look less attractive. They also make the funding risk easier to understand.

I would use unit economics to improve the offer, not only to reject spend. Better onboarding, fewer returns and more relevant customers can change the model as meaningfully as cheaper clicks.

Growth becomes a business result when the economics survive delivery.

Start with one completed customer journey

Trace one representative order or contract from acquisition through collection, fulfilment and any refund. Identify every relevant cost and its timing. Then use that verified structure to build a cohort view before deciding how much more demand to buy.