Search this question and you’ll get registration guides. Company structures, GST thresholds, Udyam, licences.
All necessary, all well covered elsewhere, and none of it is why businesses fail. Nobody has ever closed because they picked the wrong legal structure. They close because nobody bought anything.
So this is the other half — what the first ninety days actually look like from inside.
How to start a business in India
The paperwork order is: decide the structure, register, open a current account, register for GST if you cross the threshold or need it for B2B clients. That part is genuinely straightforward and a competent CA will handle it in a fortnight.
The order that decides whether the business survives is different: find the customer, sell something manually, take money, deliver it badly, fix it, then build the structure around what’s actually working.
Most first-time founders do these in the wrong sequence, and the wrong sequence is expensive.
Days 1–30: prove somebody will pay
Talk to twenty potential buyers
Not friends, not family, not people who will be encouraging. Twenty people who have the problem and control a budget.
Ask what they do about it now, what that costs them, and what they’ve already tried. Don’t pitch. You’re collecting the vocabulary your customers actually use, which becomes your marketing copy later.
Sell before you build
Offer the thing before it exists. A service you deliver manually, a pre-order, a paid pilot.
If nobody buys at this stage, you’ve saved yourself eight months and most of your capital. That is a successful outcome for month one, even though it doesn’t feel like one.
Take money from one person
The first payment changes the nature of what you’re doing. Everything before it is research; everything after it is a business with an obligation.
Do this before registration if your structure allows it. A proprietorship can invoice from day one.
Days 31–60: deliver it, badly, on purpose
Do the work manually
Whatever you eventually want to automate, do it by hand first. You’ll discover what customers actually need rather than what you assumed, and that discovery is the entire value of this month.
Automation built on assumptions is the most common expensive mistake in year one.
Get the paperwork right now, not earlier
With revenue arriving, the structure matters. This is the point for the CA, the registration, the current account, GST if applicable.
Doing it now means you register the business you actually have rather than the one you imagined in month zero — and those are frequently different.
Track two numbers only
What came in, what went out. On paper is fine.
Elaborate financial modelling in month two is a form of procrastination that looks like diligence.
Key takeaway: Registration makes a business legal. Revenue makes it a business. Doing them in that order is how people run out of money before finding out whether anyone wanted it.
Days 61–90: find out if it repeats
One sale is an accident. The question for month three is whether it happens again, predictably, without you personally convincing each person.
Where did the buyers actually come from? Usually one channel does most of the work. Find it and stop spreading yourself across five.
Can you price it properly? Most first-time founders underprice badly, then can’t afford to deliver well. Raising prices on the next customer is easier than on the current one.
What breaks at ten customers? Whatever strains at three will collapse at thirty. That’s the thing to fix now.
The Indian specifics that catch people out
Payment terms. B2B clients here frequently pay at 45, 60 or 90 days. A profitable business can run out of cash entirely on timing. Ask about payment terms before you agree a price, not after.
Family capital comes with governance. Money from relatives is cheaper and rarely free. Agree in writing what it is — loan or equity — while everyone is still pleased about it.
GST registration is often a sales requirement. Corporate clients frequently won’t onboard an unregistered vendor regardless of your turnover. That changes the timing for B2B.
Local networks matter more than they should. In most Indian cities the first ten customers arrive through somebody who vouched for you. That’s not a shortcut — it’s the actual distribution channel, and it’s worth building before you need it.
The part that decides it
Ninety days in, nearly everyone has a version of the same problem: it’s working slightly, it’s not working enough, and nobody is telling them whether to persist or change.
That’s the point where most businesses quietly end — not in a collapse, but in a gradual reduction of effort. Which is why the ability to keep going when nothing is visibly happening matters more in year one than any amount of planning, and why the habit of claiming outcomes nobody assigned you is the thing worth building before you start.
If you’re still deciding whether to start at all — including whether you want a business or a startup, which are different things — that question comes first.
And if what’s missing is someone who’s been through the ninety days and will tell you which of your problems is normal, that’s what Team United and the sessions around it are for.
Frequently asked questions
Q1. What’s the first legal step to start a business in India?
Ans. Choosing a structure — proprietorship, LLP or private limited — then registering accordingly. A proprietorship can begin invoicing almost immediately.
Q2. Do I need GST registration to start?
Ans. Only above the turnover threshold, or if clients require it. Many corporate buyers won’t onboard unregistered vendors, which often forces the decision early.
Q3. How much capital do I need?
Ans. Less than most people assume if you sell before building. Keep the first version small enough that losing it doesn’t end the attempt.
Q.4 Should I register before getting my first customer?
Ans. Not necessarily. Proving someone will pay matters more, and registering later means you register the business you actually have.
Q5.What’s the most common first-year mistake?
Ans.Building for months before selling anything, then running out of runway before learning whether there was demand.
Q6.How long until a new business becomes profitable?
Ans.Plan for eighteen to thirty-six months. Some are faster, but planning for the fast case is how people get caught out.
Q7. Can I start while employed?
Ans. Often yes, depending on your contract. Check for exclusivity and non-compete clauses before you begin taking clients.
Q8. What about loans and government schemes?
Ans. Worth exploring once you have revenue and records. Applying before you have either usually wastes weeks.
Q9. Do I need a website first?
Ans. No. One paying customer first. A website is useful once you know what you’re selling and to whom.
Q10. How do I find my first ten customers?
Ans. Almost always through someone who will vouch for you. Start with the network you have rather than the audience you don’t.