If you run an ecommerce brand in India, you already know the real silent killer of profitability isn’t ROAS, CAC, or competition.
It’s RTO (Return to Origin).
Every “delivery failed” means wasted ad spend, returned inventory, and negative margins.
Most brands try to fix RTO with tactical patches; excluding past RTO customers, building lookalikes of successful buyers, blacklisting pincodes, tweaking checkout flows, or analyzing patterns endlessly.
All valid moves. But none of them solve the root problem.
They don’t teach Meta or Google to stop bringing RTO-prone customers in the first place.
The real question isn’t how to reduce RTO after it happens. It’s how to prevent it at the algorithm level.
This is the shift most brands miss and the one that changes everything.
That’s exactly what Gifting Studio, a leading D2C brand, did. They reduced RTO by 30% without changing a single product, creative, or landing page.
Let’s break down how.
What Do Indian Ecommerce Brands Struggle With Most? Profits? Growth? ROAS?
Ask any ecommerce founder in India what keeps them up at night; it’s RTO. When 25–40% of COD orders come back, profits disappear through wasted ad spend, reverse logistics, blocked inventory, and operational chaos.
And here’s the real problem: RTO isn’t just logistics, it’s targeting. Your ads are attracting the wrong customers.
Meta and Google don’t care if the order gets returned. They optimized for “Purchase” and they delivered. Job done, from their perspective.
The Magnitude of the RTO Problem
Let’s put numbers to it.
Scenario:
Indian fashion brand spending ₹10 lakhs/month on Meta
AOV: ₹1,500 | Product cost: ₹600 | Shipping (fwd + rev): ₹150
Meta CPP: ₹300 | RTO rate: 30%
The Math
| Metric | Value |
|---|---|
| Orders generated | 3,333 |
| Orders returned (30% RTO) | 1,000 |
| Wasted ad spend on RTO orders | ₹3,00,000 |
| Wasted shipping costs | ₹1,50,000 |
| Total RTO loss/month | ₹4,50,000 |
That’s 45% of your ad budget gone and this doesn’t even include COD charges, warehouse handling, support time, or damaged goods.
The Real Problem: Your PMF Is Fine. Your Measurement Is Broken.
Your product-market fit isn’t the issue. You have happy customers, repeat buyers, and strong reviews.
The problem? Meta can’t tell the difference between a real customer and an RTO order.
To the algorithm, both look the same:
-
Click
-
Add to cart
-
Checkout
-
Purchase event fired
Meta optimized for “Purchase” and it was delivered.
Whether the customer actually accepted and paid for the product?
Meta has no idea.
And that’s where the system breaks.
Common Fixes (And Why They Don’t Solve the Core Problem)
When RTO rises, brands usually try tactical fixes:
Excluding past RTO customers
You remove known bad users but Meta keeps bringing new RTO-prone buyers every day. It limits damage, but it doesn’t change how the algorithm finds customers.
Lookalikes of non-RTO buyers
A smart move in theory but Meta’s models are still trained on all Purchase events, including RTOs. If your conversion signals are mixed, your lookalikes will be too.
These tactics help at the margins but they don’t fix the underlying targeting problem.
The Fix: How This Indian Brand Reduced RTO by 30% Using Signal Engineering + 1PD Ops
Step 1: Enable first-party data tracking
Step 2: Setup custom conversion event for signal engineering
For the gifting studio the next move was to go beyond “purchase” events and build a full set of custom signals; this is where true Signal Engineering happened.
We created:
-
Prepaid purchase, COD purchase, Normal purchase custom events
-
New Customer and Repeat Customer purchase events
Mapped these events to campaigns on both Meta and Google Ads.
This helped us review the performance of the campaigns on which is delivering prepaid, returns, CODs etc.
Here is the way you validate the quality of the data, EMQ measures how well your first-party data matches users back to Meta. A higher score (8.5+) ensures stronger attribution, better optimization, and more reliable algorithm learning.
Signal measurement to evaluate the campaign performance. Now we measure which campaign brings in more prepaid, RTO, cancel and other events like new customers , repeats etc.
The Result: 30% Reduction in RTO
After optimizing campaigns around prepaid purchase signals:
RTO dropped by 30%.
Campaign performance became measurable by signal type, prepaid vs. COD, new vs. repeat, delivered vs. returned.
The algorithm progressively learned to find buyers who complete their orders
No product changes. No new creatives. No landing page overhauls. Just cleaner data, sent to smarter systems.
If you are facing the same issue with your ecommerce brands, grab this free audit now (only limited slots available).
