Attribution · Indian DTC

Attribution is broken in India - and global tools aren't solving it

All global attribution tools (Triple Whale, Hyros, Northbeam, Cometly etc) are sitting at around 55% accuracy when used in India. These tools were built for the US and Europe and there are three specific reasons they don't work nearly as well here.

Reason 1

3rd-party checkout partners

In India, every brand uses a third-party checkout partner. They are an incredibly important tool in every D2C brand's kit. The convenience they offer is too good to pass up
→ discounts,
→ multiple payment options,
→ COD,
→ pre filled address for a quick checkout.

But there's a downside that doesn't show up on any dashboard.

Here's what a typical user journey actually looks like:

1. The user starts their journey on your Shopify store

2. They browse a few pages and click on checkout

3. Shopify hands a cart_token to your checkout partner, essentially saying: "at the end of this session, please hand this back so we know which session to link to this order"

4. 90% of the time, everything works smoothly.

5. 10% of the time, that cart_token doesn't come back and Shopify has no idea which user journey belongs to this order.

This percentage varies by checkout partner. We did a study on this. The best ones in India are losing 7–8% of sessions. The worst are losing as much as 25%.

~10%

Average signal loss from 3rd-party checkouts

Reason 2

High rate of RTOs and cancellations

Most tracking setups use the purchase event as the conversion signal. The data you optimise ads on, the data you use for retargeting. In the Indian D2C reality, that's not actually the conversion signal.

The final conversion signal is order delivery. Because even after an order is placed, roughly 15% of orders get cancelled or RTO'd.

Any attribution tool that doesn't integrate with shipping is missing those 15% signals entirely and feeding your ad platform false positives.

~15%

Signal loss from RTOs and cancellations that get miscounted as real conversions

Reason 3

Long purchase journeys

Indian consumers are much more value-conscious than their US or European counterparts. Most purchase journeys involve multiple sessions of research and price-stalking. We're talking at least a couple of weeks, sometimes a few months.

The standard 7DC1DV attribution window of 7-day clicks and 1-day views just doesn't work here. And if you're not doing cross-browser attribution and building an identity graph, you're losing the earlier sessions entirely. Especially if these earlier sessions happened on a different browser or in a Webview.

As the average number of sessions per journey goes up, so does the likelihood of cross-browser, cross-platform journeys.

33% of user journeys take longer than 7 days

8% take longer than 30 days

~25%

Signal loss from purchase journeys that fall outside the standard tracking window

So, quick recap

3rd party checkout partners → ~10% signal loss.

RTOs and cancellations → ~15% signal loss.

Longer user journeys → 20–25% signal loss.

Add it all up and a global attribution tool that doesn't solve for these three challenges is giving you an accurate picture for only 50–55% of your user journeys.