Many ad accounts run on targets that nobody can explain. A target ROAS of four because an agency suggested it, a cost per lead of a round number because it felt reasonable, or a benchmark copied from a different industry. These targets can be badly wrong in either direction: too strict, and you turn away profitable customers; too loose, and you buy growth that loses money. The right targets come from your own unit economics. This article shows how to calculate break-even ROAS, set a target CPA from customer value, adjust for returns, repeat purchases and sales cycles, and set different targets for different campaign types.
Start with gross margin, not revenue
Advertising costs must be paid from gross profit, not revenue. If a product sells for ₹1,000 and costs ₹600 to make and deliver, including product cost, packaging, shipping, payment fees and any other variable cost per order, the gross profit is ₹400 and the gross margin is 40 per cent. Every ROAS and CPA calculation should start here. A common mistake is to use the product margin before shipping, payment and fulfilment costs, which overstates the profit available to spend on advertising. For services and software, use the revenue minus the direct cost of delivering the service, such as hosting, delivery staff time or third-party costs.
Break-even ROAS
Break-even ROAS is the return on ad spend at which advertising profit is zero on the first order. It equals one divided by gross margin. With a 40 per cent gross margin, break-even ROAS is 1 ÷ 0.4 = 2.5. At a ROAS of 2.5, each ₹100 of ad spend produces ₹250 of revenue and ₹100 of gross profit, which exactly covers the ad spend. Above 2.5 the first order is profitable; below it the first order loses money. This single number tells you more than any industry benchmark, because it reflects your own margins.
Target CPA from customer value
For lead generation and subscription businesses, CPA is more useful than ROAS. Start with the gross profit a customer generates over a period you are willing to wait for payback. If a B2B customer pays ₹50,000 a month at an 80 per cent gross margin and you are willing to wait twelve months to recover acquisition cost, the gross profit over that period is ₹4,80,000. If you decide to spend up to a third of that on acquisition, your maximum cost per customer is ₹1,60,000. Work backwards through your funnel: if one in five qualified opportunities becomes a customer and one in four leads becomes a qualified opportunity, the maximum cost per lead is ₹1,60,000 ÷ 5 ÷ 4 = ₹8,000. Set the working target below the maximum to leave room for error.
Worked examples
These examples show how the same method gives very different targets for different businesses.
| Business | Inputs | Calculation | Result |
|---|---|---|---|
| D2C, first order | Order value ₹1,000, gross margin 40% | Break-even ROAS = 1 ÷ 0.40 | Break-even ROAS 2.5; target above it |
| D2C with repeat purchases | Gross profit ₹400 per order, 1.8 orders in 6 months | Gross profit per customer = ₹720; spend up to half | Maximum CPA about ₹360 |
| D2C with COD returns | Gross margin 40%, 20% of orders returned | Effective margin on placed orders falls | Break-even ROAS rises; use RTO-adjusted ROAS |
| B2B subscription | ₹50,000 a month, 80% margin, 12-month payback | ₹4,80,000 × one third ÷ funnel rates | Maximum cost per customer ₹1,60,000; per lead about ₹8,000 |
Figures are illustrative to show the method. Use your own margins, return rates and conversion rates.
Adjusting for returns and cancellations
For D2C brands in India, cash on delivery orders that are refused or returned can change the maths significantly. If a share of orders never becomes revenue, the ROAS reported by ad platforms, which counts placed orders, overstates the real return, and the shipping and handling cost of returned orders adds further cost. Calculate ROAS on delivered, paid orders, or apply a return-adjusted margin. The RTO-adjusted ROAS article and calculator on this site walk through the adjustment. Subscription businesses should make a similar adjustment for early cancellations and refunds.
Repeat purchases and how long you can wait
If customers buy again, the first order does not have to be profitable. The question becomes how long you can wait to recover acquisition cost, which depends on cash rather than preference. A business with plenty of cash and reliable repeat purchase data can accept a first-order loss and recover it within a few months. A business with little cash or uncertain repeat behaviour should aim for first-order profit or a short payback. Base repeat purchase assumptions on your own cohort data, measured over a fixed period, rather than on lifetime value estimates that assume customers stay for years.
Different targets for different campaigns
A single target across the account is rarely right. Branded search captures people already looking for you and typically shows high ROAS, much of which would have happened anyway. Prospecting campaigns reaching new audiences show lower measured ROAS but bring in genuinely new customers. Retargeting falls in between. Set targets by campaign type: stricter targets for branded search and retargeting, looser targets for prospecting that brings new customers. Then check the overall result against your blended targets using MER, total revenue divided by total marketing spend, so the account as a whole stays profitable.
Putting targets into the ad platforms
Once targets are set, use them consistently. In Google Ads and Meta, value-based bidding with a target ROAS or a target cost per acquisition only works well if the conversions you send are the ones that matter: purchases with correct values, or for lead generation, qualified leads and sales imported from the CRM. Set platform targets slightly stricter than your real limits, because platforms tend to report conversions generously. Review targets monthly against delivered revenue and quarterly against changes in margin, pricing and conversion rates. Targets that are never revisited drift out of line with the business.
Common mistakes
A few mistakes come up repeatedly. Using industry benchmark ROAS instead of your own break-even figure, which ignores your margins entirely. Using product margin before shipping, payment and fulfilment costs, which makes break-even ROAS look lower than it is. Counting revenue from orders that are later returned or refused. Assuming lifetime value from a few months of data, or from an average inflated by a small group of very loyal customers. Setting one target for every campaign, which starves prospecting and over-rewards branded search. And setting targets once and never revisiting them as prices, margins and conversion rates change. Each of these is easy to avoid once the calculation is written down and shared with whoever manages the ad accounts, whether an in-house team or an agency.
Targets for lead generation with long sales cycles
In B2B lead generation, the outcome that pays for marketing, a closed deal, may arrive months after the click. Platforms cannot optimise directly for something that slow, so set targets at several stages: a maximum cost per qualified lead and cost per opportunity derived from your target cost per customer, as in the worked example, and a target for the share of leads that qualify. Use the earliest stage that reliably predicts revenue as the platform optimisation goal, and check monthly that cohorts of leads are converting to opportunities and customers at the rates your targets assume. If conversion rates change, recalculate the stage targets. This keeps platform targets connected to the cost per customer the business can afford, even when revenue takes a quarter or more to appear.
FAQ
Divide one by your gross margin. With a 40 per cent gross margin, break-even ROAS is 2.5. Use gross margin after product, shipping, payment and fulfilment costs.
Whatever is above your own break-even ROAS by enough to meet your profit goals. Industry benchmarks ignore your margins, so they are a poor guide on their own.
Start with the gross profit a customer generates over your acceptable payback period, decide how much of it to spend on acquisition, then divide back through your funnel conversion rates.
It depends on cash and repeat behaviour. With reliable repeat purchases and enough cash, a first-order loss can be acceptable if payback is short. Otherwise aim for first-order profit.
No. Use stricter targets for branded search and retargeting, looser ones for prospecting that brings new customers, and check the whole account with MER.
Derive a maximum cost per qualified lead and cost per opportunity from your target cost per customer and your funnel conversion rates, optimise for the earliest stage that predicts revenue, and recheck the rates monthly.