Marketing10 min read
How to calculate blended ROAS across Shopify, Amazon and Flipkart
admin
Monday, 10am. Ads Manager says Meta returned 2.4. The Amazon console shows a healthy number for Sponsored Products. Flipkart's dashboard has its own figure. Add them up and the ads apparently drove more revenue than the business booked. Your founder wants one answer: for every rupee we spend on ads, what comes back?
That number is blended ROAS across Shopify, Amazon and Flipkart: total net sales from all three, divided by total ad spend on every platform. Here is how to calculate it in 5 steps, and what to strip out.
Blended ROAS is your total revenue divided by your total ad spend, across every channel, for the same period. Plain ROAS is revenue from ads divided by ad spend; blended ROAS ignores which platform claims which sale, so no attribution model is needed.
The formula:
Blended ROAS = total net sales (all channels) ÷ total ad spend (all ad platforms)
WeltPixel's guide to blended ROAS puts the key point well: blended ROAS is not more accurate than platform ROAS, it is "attribution-independent". It answers "is our ad spend paying back as a whole?" It does not tell you which campaign to cut.
For an Indian D2C brand, "all channels" usually means your Shopify store plus Amazon plus Flipkart. "All ad spend" means Meta Ads, Google Ads, Amazon Ads and Flipkart Ads.
Each platform counts its own sales, with its own rules, so the totals overlap and disagree. Summing them double-counts some orders and misses others.
Four reasons the numbers clash:
| Source | What it counts | Default window |
|---|---|---|
| Meta Ads Manager | Website conversions Meta attributes to its ads | 7-day click, 1-day engage, 1-day view |
| Shopify | Orders, credited to the last-clicked source | Last click |
| Amazon Attribution | On-Amazon sales from off-Amazon clicks (Meta, Google, email) | 14 days |
| Amazon and Flipkart ad consoles | Sales they attribute to their own ads | Varies by ad type |
So the same order can show up in Meta and Amazon, and a Meta-driven Amazon sale can show up in neither Meta nor Shopify. If you want the full story on the Meta side, read our guide on why Meta purchases don't match Shopify orders.
Pull one revenue number and one spend number for the same dates, with the same tax and returns rules on both sides. Then divide.
Here is a worked example. The numbers are illustrative, not data from any brand.
| Line | Amount |
|---|---|
| Shopify net sales (after discounts, returns and RTO) | ₹40L |
| Amazon net sales | ₹18L |
| Flipkart net sales | ₹12L |
| Total net sales | ₹70L |
| Meta Ads spend | ₹12L |
| Google Ads spend | ₹5L |
| Amazon Ads spend | ₹3L |
| Flipkart Ads spend | ₹2L |
| Total ad spend | ₹22L |
| Blended ROAS | ₹70L ÷ ₹22L = 3.2 |
Now compare that with the platforms. Say Meta claims ₹38L, Google ₹16L, Amazon ₹15L and Flipkart ₹8L. That adds up to ₹77L of "ad-driven" revenue, which is more than the ₹70L the business actually booked. That gap is overlap, not growth.
Yes. Use revenue after returns, cancellations and RTO, because an order that comes back undelivered earns you nothing. Counting it inflates blended ROAS exactly where COD is heaviest.
RTO (return to origin) is an order that ships but comes back undelivered, common with cash on delivery. On Shopify, net sales after reversals already removes refunds, but check that your RTO orders are actually marked as returned or cancelled before you pull the number.
Marketplaces need the same care. Use the net figure after returns from Seller Central and Seller Hub, not the gross order value. If you want to go one level deeper on COD, our guide to RTO adjusted ROAS shows how to adjust platform ROAS for undelivered orders.
Use blended ROAS to judge total ad payback, platform ROAS to manage campaigns inside one platform, MER for total marketing efficiency, and TACoS for marketplace ad dependence. They answer different questions, so you need more than one.
| Metric | Formula | Best for |
|---|---|---|
| Platform ROAS | Platform-attributed revenue ÷ platform spend | Day-to-day campaign and ad set decisions |
| Blended ROAS | Total net sales ÷ total ad spend | Monthly check on whether ads pay back overall |
| MER | Total revenue ÷ total marketing spend | Board and founder view, includes agency, creators, tools |
| ACoS (Amazon) | (Ad spend ÷ ad revenue) × 100 | Amazon campaign efficiency |
| TACoS | (Ad spend ÷ total revenue) × 100 | How much of marketplace revenue depends on ads |
Amazon defines ROAS as ad revenue ÷ ad spend and calls it "the inverse" of ACoS (Amazon Ads). TACoS divides ad spend by total sales, ad plus organic (Jungle Scout). MER, as Eightx describes it, is total revenue over total marketing spend across every channel (Eightx).
Indian operators are moving the same way. Inc42's piece on the end of the ROAS era in D2C names "MER, blended CAC and contribution margins" as the metrics brands now scale on.
Often, yes, and platform ROAS hides it. A shopper sees your Meta ad, searches your brand on Amazon, and buys there. Meta records a click with no purchase, so Meta ROAS looks worse than the real effect.
The best public evidence comes from geo holdout tests. In a 3-week GeoLift test published by Haus in April 2024, US brand Fresh Clean Threads found Meta drove a 23% lift in Amazon sales alongside a 21% lift in its own DTC sales. Counting the Amazon effect raised total incremental ROAS by 14% (Haus). That is one US brand, not an Indian benchmark, but the mechanism is the same for any brand selling on both.
Two ways to see it in your own data:
This is why the channel mix matters. Renee Cosmetics' cofounder Ashutosh Valani told Inc42 that about 65% of sales come from offline stores, and only about a quarter of online sales come through the brand's own website. For a brand like that, website ROAS covers a small slice of what the ads influence.
A good blended ROAS is one above your breakeven, and breakeven comes from your margin, not an industry average. Breakeven is 1 divided by your contribution margin before marketing, meaning the share of revenue left after product, shipping, payment and returns costs.
Eightx's example: at a 45% contribution margin, breakeven sits at 2.22x (Eightx). eChai Ventures makes the same point for Indian founders: a 60% margin brand can be healthy at 2.5x while a 30% margin brand needs 4x or more (eChai Ventures).
In the worked example above, a blended ROAS of 3.2 against a 45% margin clears breakeven with room. The same 3.2 at a 30% margin would barely cover ads. Margin matters more than the multiple: Kae Capital's Sunitha Viswanathan told Inc42 that a 3X ROAS can hide a real profit margin of just 5% after COGS, discounts and returns. Our contribution margin guide shows how to calculate yours, and why ROAS fails at scale covers what to track alongside it.
Blended ROAS tells you whether ads pay back overall, not which ad to scale or pause. A strong Amazon month can cover a Meta campaign that is quietly wasting spend.
Watch for three traps:
Clevrr AI is an AI performance marketer for D2C brands. It connects Shopify, Meta Ads, Google Ads, Amazon Seller Central, Amazon Ads and Flipkart, so marketplace sales sit next to website sales and ad spend in one place. When a number moves, Klev, Clevrr's AI agent, traces the cause and puts a rupee value on it, instead of leaving you to stitch 4 exports on a Monday morning.
To be clear about the gaps: quick commerce (Blinkit, Zepto, Swiggy Instamart) and Myntra are not connected today, so those sales stay outside the picture.
No. Calculate blended ROAS ex-GST on both sides, revenue and spend, so the tax does not distort the ratio. Marketplaces treat GST differently in their reports, so convert each export to ex-GST before you add them up. What matters most is consistency: use the same rule every month so the trend line stays honest.
Not quite. Blended ROAS divides total net sales by total ad spend only. MER divides total revenue by total marketing spend, which also includes agency fees, creators, tools and creative costs. MER is the wider founder view; blended ROAS isolates paid media. Track both, and expect MER to sit below blended ROAS because its denominator is bigger.
No. Blended ROAS is attribution-independent, so it shows whether total ad spend pays back, not which platform earns it. To decide between Meta, Google, Amazon Ads and Flipkart Ads, use platform ROAS for direction, Amazon Attribution for Meta-to-Amazon effects, and holdout tests where you can afford them. Then check the change against blended ROAS.
Monthly, with each month closed at the same age, such as the 15th of the following month. Returns and RTO keep arriving after the sale, so a month's figure drops as it ages. Weekly blended numbers are fine as a trend check, but compare months only when they have had the same time to settle.
Blended ROAS across Shopify, Amazon and Flipkart gives you the one number platform dashboards never will: what the business earned back for every rupee of ad spend. Build it from net sales and billed spend, ex-GST, after RTO, and close each month at the same age. Then use platform ROAS to decide where the next rupee goes.
If you want marketplace and website sales watched next to your ad spend every morning, start free with 500 credits.
Join hundreds of D2C brands using Clevrr AI to automate their growth and efficiency.
Join 500+ D2C founders using Clevrr AI to automate their operations and scale faster.
Get Started Free
