HomeField Guides › The 72-hour test
Field Guide 03

The 72-hour test: validating ideas before spending real capital

The biggest trap in starting a micro-business is spending weeks on logos, names and gear before proving a single person will pay. Three days and no money answers that question.

South Africa · Updated 26 August 2026 · Free, no signup
The short version
The situation
You have an idea for a side hustle and you are about to spend money on it — stock, tools, a logo, a website, a name.
Why it fails
Buying things feels like progress and is measurable, while asking a stranger for money feels risky and is not. So people do the first and postpone the second, sometimes for months.
What it costs
The typical first spend before a single sale — gear, branding, stock — runs R2 000 to R15 000, most of it unrecoverable if the idea has no buyer.
What to do
Write the offer as one priced sentence, put it in front of 20 real people, and count actual commitments at 72 hours. Two or more, proceed. Zero, change it or drop it — before spending anything.

What the test is actually measuring

Not whether the idea is good. Not whether people like it. One thing only: will a stranger commit something before receiving anything?

That is the only question that predicts whether the business exists. Everything else — the name, the logo, the vehicle branding, the price list design — is decoration on an assumption. If the assumption is wrong, the decoration is money you burned proving it.

Day 0: write one sentence

Before anything moves, get the offer into a single sentence containing four things:

Worked example

"For homeowners in Brackenfell — I'll pressure-wash and re-sand a standard double driveway for R350, this Saturday or next, about two hours, I bring my own water."

Everything a buyer needs to decide, in one line. No business name. No logo. No website.

If you cannot fit it into one sentence, the offer is not clear enough to test yet. That is a finding, not a delay — vagueness at this stage becomes an unsellable service later.

Day 1: put it in front of 20 real people

Twenty is the working minimum. Fewer and a zero result tells you nothing — it might just be a quiet week.

Do not build anything. No domain, no Canva logo, no business cards. If the test says yes you will have plenty of time to build; if it says no you will be glad you did not.

Day 2: ask for a commitment, not an opinion

This is where almost every informal validation attempt quietly fails. People ask "would you use something like this?" and get a warm yes, because saying yes to a hypothetical is free and saying no to a friend is awkward.

Ask for something that costs the other person something:

Counts as a yesCounts as zero
A booked date and an address"That's a great idea"
A deposit, even R50"Definitely let me know when you start"
"Come Saturday at 9""I'd probably use that"
A name and number on a waiting list, given unpromptedA like or a thumbs-up

Count honestly. The temptation to score encouragement as validation is enormous, and giving in to it is how people end up with a garage full of equipment and no clients.

Day 3: count and decide

At 72 hours, stop and read the number:

Zero is a good result

A zero at 72 hours is not a failure — it is the test doing exactly what you built it for. The failure is the version where you skipped the test, bought R8 000 of gear, and reached the same zero four months later with the money already gone.

The three ways people cheat the test

After a yes

Deliver the first job before improving anything. You will learn more in two hours of actually doing the work — what it really costs you, how long it really takes, what the client actually complains about — than in two weeks of planning. Then price the second job properly using what you learned, and keep the overhead near zero until income is steady.

Before you rely on this

This is a practical operating guide, not legal, financial or tax advice. Taking deposits, invoicing and declaring income have specific requirements in South Africa — speak to a professional once the income is regular.

Common questions

How do I know if my business idea will work before spending money?
Put a clear, priced offer in front of at least twenty real potential buyers and ask for a commitment within 72 hours. Two or more commitments means there is a buyer. Zero means the offer, price or audience is wrong — and you found out for nothing.
Why 72 hours?
Long enough for people to see a message and reply around work, short enough that you cannot fill the time with logos, names and equipment research. The deadline exists to stop you spending, not to rush the buyer.
Do I need a registered business to test an idea?
No. A test needs an offer, a price and a way to be contacted. Registration, branding and a bank account come after someone has actually paid.
What counts as a real yes?
A date in a diary, a deposit, or an address to come to. Praise and "let me know when you start" are not commitments — count them as zero.

Test it on something concrete

Blueprint 03 walks through testing a food product on real buyers before you cook at scale — the same principle, applied to something you can hold. Free, share-to-unlock, no signup.

Get the blueprint Ask a question

New guides, when there are new guides

No newsletter, no schedule, no marketing. An email only when a new field guide or blueprint goes up.

Your address is used for that and nothing else. No tracking, no third-party lists. Reply to any mail to be removed.

This is not advice. Everything here is information to help you start a side hustle — not legal, financial or tax advice. We don't promise results. Numbers and examples are illustrative and your results will differ. Your information. This site has no accounts, no cookies and no tracking. See the full terms on the home page.