Most founders get this backwards. They wait for a lightning-bolt idea, then reverse-engineer a business around it. The problem is that the ideas which arrive as a flash of inspiration are usually the ones a hundred other people also had in the shower this morning. Good ideas are rarely invented. They are noticed.
Paul Graham put it plainly in his essay on the subject: the way to get startup ideas is not to try to think of startup ideas. It is to look for problems, preferably problems you have yourself. If you force it, you get ideas that are bad and plausible-sounding, which is the worst combination because they survive long enough to waste a year of your life.
So instead of one magic method, here are seven repeatable frameworks that real founders (Indian and global) actually used. Each one is a different lens for spotting a problem worth solving. Pick two or three, run your week through them, and you will end up with a shortlist rather than a blank page.
1. Scratch your own itch
The oldest and most reliable framework. Build the thing you personally wished existed and could not find. When you are the customer, you already understand the pain, the language, and the moment of purchase, which is most of the research done for free.
Real example: Shashank Mehta spent years fighting his own weight, cycling up and down by 40 kilos three times before he was 26. Every "healthy" protein bar he bought had an ingredient list that contradicted the front of the pack. That personal frustration became The Whole Truth, a brand whose single non-negotiable rule is that every ingredient appears on the front of the package in plain language. It grew past ₹200 crore in revenue by selling honesty to people who felt exactly what he felt.
Apply it this week: Write down the last five things you tried to buy in India and gave up on, returned, or complained about to a friend. One of those annoyances is a category. The point is not to be clever, it is to be honest about what already irritates you.
2. Live in the future and build what is missing
Also from Paul Graham. Get yourself to the leading edge of a fast-changing field, live there, and the gaps become obvious. To people still in the present, those gaps look like weird niche problems. To you they look like the future arriving unevenly.
Real example: Three friends behind Sleepy Owl were coffee obsessives at a time when Indian retail offered either overpriced cafe chains or instant coffee cut with chicory. They flew to Chikmagalur, cold-called farmers, and realised India grew world-class Arabica but exported almost all of it. Cold brew, already normal in the US, did not exist here. They were simply living a year or two ahead of the mainstream Indian coffee drinker, and they built the missing product.
Apply it this week: Name one field where you are genuinely ahead of your friends (gaming, skincare science, home fitness, regional cooking). List three things that are normal in that world abroad but absent in India. That gap is your candidate.
3. Category arbitrage (a Western model, adapted to India)
You do not have to invent the idea. You can import a business model that already works in a more mature market and adapt it to Indian pricing, taste, and distribution. This is not copying, it is arbitrage on timing. The model is proven, so your risk drops to execution and localisation.
Real example: boAt looked at a global truth (people wanted audio gear that was stylish, affordable, and bass-heavy) and rebuilt it for Indian wallets and Indian taste, sitting deliberately between cheap no-name gadgets and overpriced global brands. The category existed worldwide. The India-shaped version did not, until they made it.
Apply it this week: Pick a D2C category that is huge in the US or Europe (subscription vitamins, functional beverages, refillable home care). Ask what would have to change for it to work at Indian price points and habits. If the answer is "a lot but not everything," you may have found a wedge.
4. Unbundle an incumbent
Take one broad, mediocre, do-everything player and rip out a single slice they serve badly. Do that one thing ten times better. This is the famous "unbundling of Craigslist" observation from Andrew Parker at Spark Capital, where Airbnb, Uber and dozens of others each took one Craigslist category and turned it into a great standalone product.
Real example: In personal care, the giant incumbents sold every man an expensive multi-blade razor through crowded retail shelves. Dollar Shave Club unbundled exactly one job (getting decent blades cheaply, on a schedule) and delivered it direct, taking real market share before a billion-dollar exit. In India, the same logic plays out every time a focused brand pulls one product line out from under a sprawling FMCG portfolio and does it with more care.
Apply it this week: Pick one large, unloved incumbent you buy from (a big FMCG brand, a bloated marketplace). Write out everything they do. Circle the one thing they do worst. That circle is your product.
5. Ride a trend or a regulatory shift
Some doors open only for a few years. A new regulation, a new tax rule, a sudden cultural shift, or a technology hitting mass adoption creates a temporary tailwind. Founders who move early get demand that feels effortless because the wave is doing half the pushing.
Real example: Mamaearth rode the clean-label parenting wave. As Indian parents grew wary of harsh chemicals and started reading labels, Varun and Ghazal Alagh (who talked to hundreds of parents first) built a toxin-free range aimed squarely at that rising anxiety. The trend created the demand. They built the transparent brand that met it.
Apply it this week: List three shifts happening right now (a new government scheme, a rule change in food labelling or packaging, a habit that changed after the pandemic). For each, ask what product people will need more of because of that shift. Trends have expiry dates, so bias toward the ones just starting.
6. Boring but profitable (the repeat-purchase angle)
Glamour is a tax. The most durable D2C businesses are often built on unsexy consumables that people buy again and again without thinking. High repeat rates mean you earn back your acquisition cost once and then keep the customer almost for free, which is the whole game in D2C economics.
Real example: Country Delight sells milk. There is nothing exciting about milk, except that once a household subscribes to daily delivery, the repeat cost of acquisition is effectively zero and the revenue stream is close to indefinite. Founders Chakradhar Gade and Nitin Kaushal found their wedge in a boring truth (a large share of India's milk was adulterated) and turned trust plus routine into a business serving millions of households.
Apply it this week: Make a list of things your household reorders every week or month without deliberation (coffee, cleaning refills, pet food, supplements). Boring and repeated beats exciting and one-off. A validation-first system helps here, because it forces you to test whether people will actually reorder before you build the supply chain around it.
7. Solve a problem you keep seeing in a community
You do not have to have the problem yourself. If you sit inside a community (a subreddit, a WhatsApp group, a professional network, a hobby forum) and the same complaint surfaces again and again, that repetition is a signal. The community is doing your market research out loud.
Real example: Shashank Mehta again, but for a different reason. Before The Whole Truth sold a single bar, he spent years running a fitness blog and community called FITSHIT, watching tens of thousands of readers ask the same questions and voice the same frustrations about "health" food. That community both revealed the problem and became his first customer base, which is why his launch did not need convincing.
Apply it this week: Join or revisit one community you already belong to. Scroll for the complaint that shows up in a dozen different threads worded a dozen different ways. When many people describe the same pain independently, you are looking at demand, not a one-off gripe.
From idea to something worth building
Seven lenses, and you do not need all of them. Run your week through two or three, and you will move from a blank page to a shortlist of real problems attached to real people. That is the entire job of idea generation. It is not about a stroke of genius, it is about noticing what is already broken and who is already annoyed.
The trap comes next. Falling in love with the first idea on your shortlist and building for six months before talking to a single stranger. A validation-first system exists to prevent exactly that. Before you build, you confirm the problem is real, painful, and something people will pay to fix. Pick your idea with these frameworks, then earn the right to build it by proving demand first.
Sources
- Paul Graham, How to Get Startup Ideas: https://www.paulgraham.com/startupideas.html
- Y Combinator Library, Essays by Paul Graham: https://www.ycombinator.com/library/carousel/Essays%20by%20Paul%20Graham
- CB Insights, Unbundling Craigslist: https://www.cbinsights.com/research/craigslist-unbundling/
- The Strategy Story, Dollar Shave Club Business Model: https://thestrategystory.com/2021/07/05/dollar-shave-club-business-model/
- Mamaearth, Our Story: https://mamaearth.in/our-story
- Country Delight, Our Story: https://countrydelight.in/story
- Inc42, Inside Sleepy Owl Coffee's Growth Journey: https://inc42.com/startups/inside-at-home-coffee-brand-sleepy-owl-coffees-%E2%82%B9100-cr-growth-journey/
- The Hot Startups, The Whole Truth Case Study (Shashank Mehta): https://www.thehotstartups.com/p/the-whole-truth-foods-business-startup-case-study-shashank-mehta
