How to validate a business idea in 48 hours

You don't need a product, a brand, or a website to find out whether an idea has a pulse. You need one concrete offer, ten real people, and the discipline to count actions instead of compliments. Here's the 48-hour version, step by step.

First, know what validation actually is

Validation is evidence that someone with the problem will take a costly action to solve it: pay money, put down a deposit, book an appointment, sign up with a real email and show up. Everything else — likes, kind words from friends, survey answers, “I'd definitely buy that” — costs the other person nothing, so it proves nothing.

Hour 0–2: write down the riskiest assumption

Every idea rests on one assumption that kills it if false. Usually it's one of these:

• Pain: “People are actively annoyed enough by this to act.”
• Reachability: “I can actually get in front of the people who have this problem.”
• Price: “They'll pay enough for this to be a business, not a favor.”

Pick the one you're least sure about. Your 48 hours test that, not the idea in general.

Hour 2–6: turn the idea into one concrete offer

Vague ideas can't be validated. “I help people with their careers” gets polite nods; “a fixed-price, 6-week career-pivot sprint for stuck professionals — €590, four sessions, you leave with a decision” gets a yes or a no. Write your offer as one sentence:

Put a real price on it. A price is itself part of the test — offers without prices generate fake enthusiasm.

Hour 6–24: put it in front of 10 right-fit people

Not an ad campaign. Not a landing page with zero traffic. Ten direct, personal messages or conversations with people who plausibly have the problem: past colleagues, the community group, the subreddit, the WhatsApp group of parents, the trade association, your existing customers if you have any. The message is short and honest:

• Name the problem you've noticed.
• State the offer in the one sentence you wrote, price included.
• Ask for the action: “Want one of the first spots?” / “Can I pencil you in for next week?”

If you can't name ten reachable people, that's your answer already: the reachability assumption failed, and no website would have fixed it.

Hour 24–46: count actions, not opinions

Track three numbers only: how many saw the offer, how many replied, how many took the costly action (paid, booked, deposited, signed up). Resist upgrading kind words into signal. A useful reading guide for ten right-fit people:

• 2–3+ costly actions: strong signal. Take the money seriously — deliver, learn, raise the price later.
• Replies and questions but no commitments: the pain is real but the offer is off — usually price, scope, or trust. Change one variable and re-run.
• Silence: wrong audience, vague offer, or weak pain. Change exactly one and re-run before giving up.

Hour 46–48: decide in writing

Three honest outcomes: go (you have commitments — deliver them before building anything else), pivot (signal says the problem is real but the offer needs to change), or kill (no pain, no reach — better to know now, for €0, than after six months and a logo). Write the decision down with the numbers next to it; future-you will try to renegotiate.

What people get wrong

• Building first. A website is not validation; it's decoration on an untested assumption.
• Asking friends. They love you. They are the noisiest possible data source.
• Surveys. People are terrible at predicting their own buying. Watch behavior, not intentions.
• Testing the whole business. 48 hours tests one assumption. Serial small tests beat one giant launch.

Common questions