Why I Will Never Build For India First

There's a sentence that ends most fundraising conversations about consumer products in India before they really start, and it usually gets said gently, but it always means the same thing. Great usage numbers, but what does monetization look like. That question isn't really about the product. It's about a much harder, less comfortable fact that most founders here learn only after they've already built something people love, that loving a product in India and paying for it are two almost entirely separate behaviors, and the gap between them is wide enough to sink a company that didn't plan around it from day one. Pocket FM is the cleanest illustration of this gap I know of, because the company didn't have a demand problem, it had a geography of revenue problem. The audio drama platform built a massive, genuinely engaged user base in India, and by most accounts the Indian audience makes up a large majority of total listenership. But when you look at where the actual money comes from, the picture flips almost entirely. Western markets, the US in particular, a much smaller slice of total users, end up contributing the overwhelming majority of revenue, reportedly somewhere in the range of seventy percent from a user base that's a fraction of the size of the Indian one, roughly fifteen percent of the audience, by some estimates, generating close to seventy percent of the money. That's not a rounding error or a temporary imbalance while a market matures. That's the entire business model quietly telling you where it actually survives, and it's telling you that despite India being the loudest, most active part of the platform, day after day, in terms of listens, shares, and time spent. What's happening right now with OpenAI and Google in India makes the exact same point, from the other direction, at a scale that's hard to ignore. In November 2025, OpenAI made its ChatGPT Go plan, its most affordable paid tier, completely free for every user in India for a full year, a plan that normally costs 399 rupees a month everywhere else it's sold. This wasn't a small regional promotion, it was reportedly OpenAI's first time offering that tier free anywhere in the world, and the reason wasn't generosity, it was reportedly disclosed that ChatGPT had seen millions of downloads in India but in app purchase revenue that barely registered against that download volume, a company with a genuinely massive Indian audience discovering that almost none of it was converting into money. Around the same time, Google partnered with Reliance Jio to give eligible users an eighteen month free subscription to its AI Pro plan, a package including Gemini 2.5 Pro and premium tools, valued at roughly 35,100 rupees, again something Google hadn't done at that scale in any other market. Two of the most valuable technology companies on earth, both looking at the same market, and both arriving at the identical conclusion, that the fastest way to win Indian users is to stop asking them to pay at all, at least for now. That single fact should end most debates about whether India is a hard market to monetize directly. If OpenAI and Google, companies with essentially unlimited resources and pricing sophistication, teams full of people whose entire job is figuring out exactly how much a market will bear, concluded that giving away a premium product for free was the only viable path to real adoption in India, that's not a data point a smaller company gets to argue with. It's confirmation of exactly the pattern this piece keeps returning to, arrived at independently, by two competitors, in the same few months. Spotify's India numbers tell a similar story from a different angle. India is one of Spotify's largest markets by user count anywhere in the world, and one of its lowest by revenue per user, low enough that India is widely understood inside the streaming industry as a scale market, not a profit market, a place you're present in for the user numbers on an investor slide, not for the actual economics behind them. Netflix ran directly into this wall and had to publicly reverse course because of it. The company entered India at a price point calibrated closer to its Western markets, treated India like every other territory, and grew slowly, well behind projections, while local competitors with far cheaper, ad supported or mobile only plans pulled ahead in raw subscriber counts. Netflix eventually cut its India pricing dramatically, in some tiers by more than half, essentially admitting that the product people wanted was the same, but the price people were willing to actually pay for it was a completely different number than what worked everywhere else, and no amount of brand strength closed that gap on its own. Gaming shows the same divide even more starkly. Indian mobile gaming has enormous download and engagement numbers, some of the highest in the world, but in app purchase revenue per user lags far behind Western and Japanese markets, and the games that have actually built durable Indian revenue, Dream11 being the clearest example, did it through a real money fantasy sports model built around cricket, not the cosmetic microtransaction model that funds most Western gaming businesses. The audience size was never the problem in any of these categories. The willingness to pay for the same kinds of digital goods Western users pay for casually simply wasn't there at the same scale, and no amount of product polish on its own has closed that gap yet, for anyone, at any size. The instinct is to explain all of this away as a purchasing power gap, and that's real, genuinely real, a subscription priced comfortably for a US wallet is a meaningfully bigger ask relative to income across most of India. But purchasing power alone doesn't fully explain the pattern, because plenty of Indians spend freely on things they've decided matter, a flagship phone bought on EMI, a wedding, a weekend trip with friends. The deeper issue is that digital content and digital services specifically got trained, over more than a decade, to be free, or close to it, and every giveaway like the ones above, however good for adoption in the short term, adds one more year to that training. Piracy filled the gap before legal streaming ever arrived. Free to play mobile games normalized spending nothing and tolerating ads instead. And now, the two biggest AI companies in the world are teaching an entire generation of Indian users, at exactly the moment they're forming their first habits around this category, that premium AI access is something you get for free, not something you pay for. Whatever pricing conversation an Indian AI company tries to have with users three years from now is going to be fought against that exact memory, a memory that OpenAI and Google themselves just spent real money helping create. This is the actual, uncomfortable reason I won't build a consumer product in India expecting Indian revenue to carry it, and it's also why, if I ever do build for a paying Indian base specifically, I'm not interested in chasing hundreds of millions of free users who never convert. I would take one hundred thousand people who actually pay over five hundred million who never do, because a business built on the second number isn't a business, it's a very expensive engagement metric, and I've watched two of the best resourced companies in the world spend real money just to prove that point rather than fight it. A hundred thousand paying users is a number small enough to sound unambitious next to download counts in the hundreds of millions, and it is also the only one of those two numbers that has ever actually kept a company alive.