Validate a startup idea for under ₹5,000

Most first-time founders spend six months and their savings building a product nobody asked for. The expensive part was never the code. It was the assumption that people wanted it. You can test that assumption for the price of a nice dinner.

Validation is not asking friends if your idea is good. Everyone says yes to be kind. Validation is finding evidence that strangers will trade money, time, or effort to get what you are offering — before you build the thing.

The only question that matters

Will someone pay for this? Every cheap test below is a different way of asking that one question. Interest is free; money is expensive. When someone hands over a card number, a deposit, or a signed intent, they are telling you the truth in a way a survey never will.

Rank your evidence by cost to the person giving it:

• A "like" or a survey tick costs nothing. Ignore it.
• An email signup costs a little attention. Weak signal.
• A pre-order deposit costs money. Strong signal.
• A paid pilot costs money and time. This is a customer.

Chase the bottom of that list.

Test 1 — The one-page smoke test (₹0–₹800)

Put up a single landing page describing the product as if it exists. State the problem, the promise, and one clear call to action: "Start free" or "Pre-order ₹499." Send 100–200 real visitors to it — a small ad budget, a relevant community, a cold-outreach list.

You are measuring the click on that button, not the pageview. A 2–3% conversion on a genuine buying action from cold traffic is a real signal. Under 1% and the promise is not landing.

Test 2 — Concierge delivery (₹0)

Before you automate anything, deliver the outcome by hand. Selling an AI-powered meal planner? Make the plans yourself in a spreadsheet and email them to ten people. Charge for it. If nobody pays you to do it manually, no amount of automation changes that. If they do, you now know exactly what to build and what to skip.

Test 3 — The pre-sale (₹0)

Ask for money before the product ships. A deposit, a discounted lifetime deal, a paid waitlist. The classic version is a founder who collected 100 pre-orders for a hardware product before manufacturing a single unit. The pre-sale funds the build and proves the demand in the same move.

If asking for money feels premature, that discomfort is the point. It is the exact moment the market tells you whether it believes you.

Test 4 — Fake the feature (₹0)

Add the button for a feature you have not built. When someone clicks it, show a "coming soon — join the list" message and count the clicks. This measures real intent inside a real product far better than a standalone survey. Use it sparingly and be honest that it is early.

Test 5 — Manual outreach to 20 real buyers (₹0)

Find twenty people who have the problem and talk to them. Not "would you use this?" — that invites polite lies. Ask what they do today, what it costs them, and what they have already tried to fix it. If they have paid for a bad workaround, you have found a bleeding wound worth stitching. If they shrug, the pain is not sharp enough.

Test 6 — The competitor's ledger (₹0)

If competitors exist, that is good news — someone has proven people will pay. Read their reviews, their support forums, their refund complaints. The one-star reviews are a free list of unmet needs. Your wedge is the gap between what they promise and what customers keep complaining about.

Reading the results

Run two or three of these in parallel over two weeks. You are looking for one thing: a repeatable action where a stranger gives up something valuable to get what you offer. One paying pilot customer beats a thousand survey responses.

If the tests come back cold, you have not failed — you have saved six months and your savings. Change the promise, change the audience, or change the idea, and run the cheap test again. That loop, run fast and honestly, is the whole job in the early days.

The build comes after the evidence, not before it.