By Sanjay Trisal, GM, India, AppsFlyer
September 16, 2026: Every festive season, finance marketers converge on the same seven days. Diwali gets the media plan, the war room, and the results deck circulated to leadership afterward. Our numbers say that week doesn’t decide the season.
AppsFlyer’s India Festive Report 2026 tracked 2.1 billion installs, 5 billion re-engagement conversions, and $700 million in ad spend across India’s 2025 festive window. Inside that scale sits a pattern most finance marketers never check for: the vertical’s biggest revenue week didn’t fall during Diwali. It landed the week after, alongside Chhath Puja, and pulled in 13% of the season’s Android revenue and 11% of iOS revenue, ahead of Diwali week’s own 11% and 9%.
The spend backs up the shift. Re-engagement budgets rose 102% on Android during Diwali week, then climbed another 97% the week after. Conversions moved even more, up 124% on iOS and 23% on Android over that same stretch. Read on its own, this looks like the best kind of marketing story: customers who came back for Diwali stayed engaged well past it. A loyalty win, hard-earned.
I’d have believed that story too, if I’d stopped at the top-line numbers. Retention data tells a sharper story. iOS cohorts acquired in this window hit 13.1% retention on Day 7, well above the usual 11.5%. By Day 30, that lead has shrunk to 8% against a baseline of 7%, a gap that’s nearly closed. Android shows almost none of this movement, at Day 7 or Day 30. Average the two platforms together, and the whole pattern vanishes. What looks like sustained loyalty is really a short, real, platform-specific window, not a lasting behavior change.
October’s install data points to the same conclusion from a different angle. The events driving the post-Diwali surge – gold buying around Dhanteras and financial planning tied to Chhath Puja – also produce the strongest install-to-purchase conversion of the entire year: 3% on Android, 7% on iOS. This isn’t a single-week sentiment spike. It’s intent that shows up consistently, in a season built around exactly this kind of financial decision-making.
None of this shows up on a chart that plots one metric against time. It only appears when spend, conversion, revenue, retention, and platform get read together, and even then only if nobody averages away the split between Android and iOS. That’s the harder discipline, and it pays off.
I don’t think the answer is to wait until Day 30 to decide whether a spike was real, and I don’t think it’s to write off the week after Diwali as a short transaction bump dressed up as engagement. The value sits earliest, in the days right after Chhath Puja, while that financial intent is still active. For finance marketers in India, that week is this season’s clearest opening. It rewards whoever measures it precisely enough to see it: split by platform, counted in the week it’s actually earned, not folded into an average that hides it.