CPM, CTR, Conversion Rate, ROAS and ROI Explained (With Benchmarks)

CPM is what 1,000 ad impressions cost, CTR is the share of people who click, and conversion rate is the share of visitors who buy. Multiply them through and you get cost per order, which you compare with the profit each order brings in. If an order costs less in ads than it earns before ads, the campaign makes money. You can check each number with the free CPM calculator, CTR calculator and conversion rate calculator.

This guide works through every metric, including ROAS, ROI and break-even ROAS, using one realistic Indian D2C campaign in rupees. All figures from the calculators were produced by running the site's own calculator code. Benchmarks come at the end, with their sources, because your own past campaigns are a better yardstick than any industry average.

The formulas

"Contribution margin" is the share of each sale left after the costs that come with every order: the product, packing, shipping, payment fees and returns. It is not the same as your markup, a difference explained in our post on margin vs markup.

Worked example: one D2C campaign, step by step

A small Indian brand sells a gift box for ₹1,180 including 18% GST, so ₹1,000 is its own revenue and ₹180 goes to the government. Each order costs it ₹500 to fulfil: ₹300 for the product and packaging, ₹90 for shipping and ₹110 set aside for payment fees and returns. It runs an Instagram and Facebook campaign for a month. Here is what Ads Manager and its analytics show:

• Ad spend: ₹60,000
• Impressions: 5,00,000
• Link clicks: 6,000
• Landing-page sessions in analytics: 5,000
• Orders: 150

Step 1: CPM. ₹60,000 ÷ 5,00,000 × 1,000 = ₹120. The CPM calculator shows 120.00, and ₹0.12 per impression.

Step 2: CTR. 6,000 ÷ 5,00,000 × 100 = 1.2%. Use the link-click figure, not all clicks: Meta reports both "CTR (all)", which counts every kind of click on the ad, including likes and comments, and "CTR (link click-through rate)", which counts only the clicks that send people to your site.

Step 3: CPC. ₹60,000 ÷ 6,000 = ₹10 per click. Check: ₹120 CPM ÷ (1,000 × 1.2%) = ₹10.

Step 4: conversion rate. 150 orders ÷ 5,000 sessions × 100 = 3%; the conversion rate calculator also shows 33 visitors per order. Why only 5,000 sessions from 6,000 clicks? Some people tap and leave before the page loads, and some visits are not recorded because of blocked trackers or declined cookies. Measured against clicks instead, 150 ÷ 6,000 = 2.5%. Use one definition consistently; this example uses the 2.5% click-to-order rate from here on, because clicks are what you pay for.

Step 5: CPA. ₹60,000 ÷ 150 = ₹400 per order. The whole chain in one line: CPA = CPM ÷ (1,000 × CTR × click-to-order rate) = 120 ÷ (1,000 × 0.012 × 0.025) = ₹400.

Step 6: ROAS. Ads Manager reports the purchase value your website sends it, which often includes GST, depending on how your store is set up: 150 × ₹1,180 = ₹1,77,000, a ROAS of 2.95. Your own revenue is 150 × ₹1,000 = ₹1,50,000, a true ROAS of 2.5. Know which one you are looking at.

Step 7: break-even ROAS. Each ₹1,000 sale leaves ₹500 after fulfilment, a contribution margin of 50% (the profit margin calculator shows a 50% margin and a 100% markup for cost ₹500, price ₹1,000). Break-even ROAS = 1 ÷ 0.50 = 2.0 on revenue excluding GST, or ₹1,180 ÷ ₹500 = 2.36 on the GST-inclusive figure Ads Manager shows. The campaign's 2.5 (or 2.95) is above break-even, so it made money. In cost-per-order terms, the break-even CPA is the ₹500 contribution; the campaign paid ₹400.

Step 8: ROI. Contribution before ads: 150 × ₹500 = ₹75,000. In the ROI calculator, enter ₹60,000 as the investment and ₹75,000 as the return: ROI 25%, net gain ₹15,000. A common mistake is entering revenue as the return. With ₹1,50,000, the same calculator says 150% ROI and a gain of ₹90,000, which ignores the cost of the products you shipped.

Which lever to pull: a what-if table

Because CPA is CPM divided by the product of CTR and click-to-order rate, a change in any one of them moves the cost of an order. Here is the same ₹60,000 budget with one lever changed at a time:

This turns diagnosis into a routine. If cost per order rises, check which of the three changed. CPM up: the audience got more competitive or narrower. CTR down: the creative is tired or aimed at the wrong people, so rotate it before raising the budget. Conversion down with CTR steady: the problem is after the click, such as page speed, price, delivery charges or a confusing checkout. Fix one thing at a time so you can tell what worked.

Does the campaign cover the business?

A campaign can beat break-even ROAS and still leave the business short, because rent, salaries and software are not in any of these numbers. Say this brand's fixed costs are ₹1,50,000 a month. In the break-even calculator, enter fixed costs of ₹1,50,000, a price of ₹1,000 and a variable cost of ₹900 (₹500 fulfilment plus ₹400 of ads per order): it needs 1,500 orders a month, ₹15,00,000 in revenue, because each ad-driven order contributes only ₹100. Orders that arrive without ads, from repeat customers, search or word of mouth, contribute the full ₹500, and with a variable cost of ₹500 the calculator shows 300 such orders would cover the fixed costs.

That is why repeat purchases matter so much. If a share of these 150 customers buys again without being re-acquired through ads, the true return on this month's spend is higher than the 25% above. Track it from your order data, not from Ads Manager.

Benchmarks, and why to use them carefully

Published benchmarks give a rough sense of normal, but they vary widely by industry, country, objective and season, and most public datasets are not Indian. The most widely quoted ones come from WordStream by LocaliQ, a US-based agency, based on its clients' campaigns, with costs in US dollars:

The same reports give an average cost per click of $5.42 for search ads and $0.60 for Meta traffic campaigns. Treat all of it with care. Search ads catch people already looking for something, so they naturally get higher CTRs than social ads that interrupt a feed. A "conversion" in these reports is whatever each advertiser counted, often a lead form or a call rather than a purchase, so e-commerce sales rates can be much lower. And dollar costs from US campaigns tell you little about rupee costs in India. We could not find a public, regularly updated benchmark for Indian campaigns, so the most useful benchmark is your own average over the last three months, broken down by campaign type.

Measure it properly

• Tag every ad link with the UTM link builder so your analytics, not just the ad platform, can attribute sessions and orders to each campaign.
• Compare platform-reported purchases with orders in your store. Ad platforms count conversions under their own attribution windows, and the numbers rarely match exactly.
• Count all costs when you judge ROI, including agency fees, influencer payments and design work, or every campaign looks better than it was.

Frequently asked questions

What is a good CPM in India?

There is no single figure. CPM depends on the platform, the audience's size and buying power, the placement and the season, and it rises when many advertisers target the same people. Judge CPM by the cost per order it produces, not on its own: a higher CPM for a buyer-ready audience can give a lower CPA.

What is the difference between ROAS and ROI?

ROAS divides revenue by ad spend and ignores every other cost. ROI subtracts the ad spend from the profit the sales generated, then divides by the spend. A ROAS of 2.5 can be a 25% ROI, as in the example, or a loss if your margins are thin.

How do I calculate break-even ROAS?

Divide 1 by your contribution margin. With a 50% margin it is 2.0; with 25% it is 4.0; with 20% it is 5.0. Use revenue without GST, or convert the target, if your Ads Manager purchase values include GST.

Is conversion rate measured on clicks or on visitors?

Both are used. Ad platforms usually divide by clicks; analytics tools divide by sessions or visitors, which are fewer. Pick one and use it every time, and label it in reports.

Why is my CTR high but sales low?

The ad is persuasive but the page does not deliver: it may be slow, the price or delivery charge may surprise people, or the ad may promise something the page does not show. Test the page on a mid-range phone on mobile data, and make sure the offer in the ad appears at the top of the page.

Sources: WordStream by LocaliQ, 2026 Search Advertising Benchmarks (updated June 2026) and Facebook Ads Benchmarks (updated September 2026); Meta Ads Manager metric definitions. Calculator results from the site's CPM, CTR, conversion rate, profit margin, ROI and break-even calculators.