How to Test an Idea Cheaply Before You Build:Five Ways

To test an idea cheaply before you build, pick the least expensive thing that could genuinely change your mind: a conversation, a page with a price on it, a rough prototype, doing it by hand for one person, or data you already hold. Then decide in advance what a no looks like.

How to test an idea cheaply before you build

Building is the most expensive test you own.

It feels like the natural way to find out. You build it, you see what happens. But for a one-person business the price isn't money. It's evenings. And evenings are the only budget that really matters, because you can't hire more of them. Every question you answer by building is a question you could have answered for a fraction of the cost. Discovery exists to do exactly that: answer cheaply what would otherwise take weeks.

So the rule is blunt. Before you build anything, ask what the cheapest thing is that would move your belief. Then do that instead.

The five cheap tests

You don't need a lab. You need one of these, matched to the question you're asking.

Look at what's missing from that list. Code. A launch. A polished anything.

The fourth row deserves a second look. Doing the work manually for a single person is the most underrated test of the lot. If you're thinking about automating a service, deliver it by hand once. You'll learn what the customer values, what actually takes the time, and whether they'd have paid. All before you've written a line of anything.

Free guide · 21 pagesDiscovery and delivery, one person.The Dual-Track Playbook: how to keep discovering while you keep building.Get the free guide →

Match the test to the question

Different questions need different tests. A common mistake is using a good test for the wrong question.

A conversation can tell you whether a problem is real. It can't tell you whether people will pay, because people are generous in conversation and much stingier with their time and money. A small commitment beats a large compliment every time. That is why a page with a price on it exists: it asks for a step, not for an opinion. A prototype can show whether people understand your idea. It can't tell you whether anyone cares, because curious people click on things. So start with your riskiest assumption, the one you found in finding your riskiest assumption, and pick the test that speaks to that one. Not the one you find most enjoyable to build. People reliably choose tests they like.

The only question that matters before you start

Stop and ask yourself something uncomfortable.

If this comes back negative, will I actually stop?

Be honest. If the answer is no, don't run the test. Build the thing. At least that's an honest decision. A test you'd ignore is worse than no test at all, because it costs you an evening and manufactures false confidence. You come out of it feeling validated when all you did was repeat what you'd already decided.

If the answer is yes, write down what result counts as a no. Do it before the test, not after. An unstated bar always moves to fit what you wanted to do. That's not a character flaw. It's how people work. So take the choice away from your future self.

Put a clock on how you test an idea cheaply

Every test needs a deadline. A few days, a week.

Discovery fills whatever time you give it. It always feels responsible to learn a little more. One more conversation, one more variation of the page, one more week for data to accumulate. That impulse is how a cheap test becomes an expensive project. Set the limit before you begin. When the date arrives, make a decision with what you have. Three possible decisions: build, stop, or "I still don't know and here is the one thing I'll test next." All three are legitimate. Staying in limbo is not.

A small example of how to test an idea cheaply

Say you run a one-person design studio and you want to launch a monthly retainer for existing clients. Your riskiest assumption: clients will commit to a recurring payment rather than keep buying one-off projects.

The expensive test: build the whole retainer service, packaging, onboarding, the lot, then offer it.

The cheap test: write a short description with a price, and offer it to a handful of past clients who match the type you want. You decide in advance that if nobody takes a step, you'll shelve it. You give yourself a week.

Either someone says yes, or the idea saves you a month. You've spent an afternoon, not an autumn.

Notice what the cheap version needs: a clear question, a price, a bar, and a date. Not skill. Not tools.

Free assessmentHow trapped are you, really?Take the Corporate Suffocation Index, the free solopreneur readiness assessment.Take the assessment →

Why scaling makes this more important

When you want to grow without hiring, every build decision is a bet with your own time. A team can absorb a few misses. You can't. One wrong month is a month of the only labour you have.

That changes how you should think about "small" bets too. Adding a feature, a second offer or a new channel looks cheap because it's only a weekend. But AI has made the weekend so cheap that the real cost has quietly moved. It's no longer the effort of building. It's the opportunity: everything else you could have done with that weekend.

Cheap tests protect that. They turn an expensive bet into a small one, and they give you a way to say "not yet" without feeling like you're stalling. The question you can't answer at your desk gets a date and a method. You'll know by Friday.

There's a second benefit, quieter and easy to miss. Each test teaches you something about your own judgement. You start to notice which of your hunches survive and which don't. That calibration is worth more than any single result, and you only get it by testing often and small.

What cheap tests do not replace

Be careful. Cheap is not the same as sloppy.

Talk to people who actually match your target, not whoever happens to be available. Your friends are available. They're also polite. Three friendly chats prove nothing, and neither does a poll where everyone said yes.

And ask them the right way. Ask about behaviour and history, not about whether they'd like your idea. We cover that in how to talk to customers without leading them. A cheap test run badly is still a bad test.

Finally, don't confuse "cheap" with "never build." Building is how an idea that survived testing gets checked against reality. The goal isn't to avoid it. The goal is to arrive at it with evidence.

Pick your test on Monday

Here is the practical version. Take the one assumption you most fear is wrong. Write it as a plain sentence about a customer. Next to it, write which of the five tests speaks to it. Then write the result that would count as a no.

Three lines. Then book the time, because a test with no slot in the calendar is a wish. If the test needs other people, send the message today, while you still feel a little uncomfortable about it. That discomfort is a decent sign you picked the right assumption.

When the date arrives, look at what happened and make the call you promised yourself. No fresh rounds of "just one more." You wrote the bar when you were calm. Trust that version of you.

Free guide · 21 pagesDiscovery and delivery, one person.The Dual-Track Playbook: how to keep discovering while you keep building.Get the free guide →

Write down what you learned

Four lines will do. What I asked. What I did. What I found. What I decided.

It costs nothing at the time of writing. And it stops you relearning the same thing next year, or running the same test on an idea you already killed. When you do arrive at the point where you have enough evidence to commit, deciding when an idea is validated is the next question to settle.

Free assessmentHow trapped are you, really?Take the Corporate Suffocation Index, the free solopreneur readiness assessment.Take the assessment →

Frequently asked questions

What is the cheapest test I can run?

Usually a conversation or looking at data you already have. Both cost almost nothing. Choose the one that speaks to your riskiest assumption, not the one that is simply easiest.

Is a landing page with a price a real test?

It's a good one for the question "will anyone take a step towards paying?" because it asks for a small commitment instead of an opinion. It isn't a test of whether the problem is real. Use a conversation for that.

What if I would build it anyway?

Then skip the test and build. A test you would ignore costs an evening and creates false confidence. Running it only makes sense if a negative result would genuinely stop you.

How long should it take to test an idea cheaply?

Set a limit before you start, usually a few days to a week. At the end, decide with what you have, even if the decision is to name one more thing to test. That is how you test an idea cheaply and keep it cheap.

The full method, with the checklists, is in The Dual-Track Playbook, a free 21-page playbook.