Guide
How to test a business idea before you build it
A practical method for a real answer in about a week, for around €100 of ads. It is the same method we use for our clients.
In short
- Ask strangers, not friends. Real interest is a click, a sign-up or a refundable pre-order.
- Write down your pass mark before you start, so the numbers decide.
- Build one page with one offer, one price and one step to take.
- Send 100 to 400 targeted visitors from the channel your buyers already use.
- Compare the result with the pass mark: go, rework or stop.
Why friends and surveys don’t count
People who know you are kind, and everyone is generous with an imaginary wallet. Saying “I’d buy that” costs nothing. What you need is behaviour: a stranger sees your offer, with the price on it, and takes a step. That is the signal you can bet months of work on.
The method, step by step
- Write the offer in one sentence. Who it is for, what they get and what it costs. For example: “Birthday cakes for offices, delivered by 10 am, €39 for a team of 10.” If you can’t write it yet, you aren’t ready to test it.
- Set the pass mark. Before any ad runs, write down the result that would make you build, such as “8% of visitors join the waitlist” or “five refundable pre-orders”. Deciding first stops you explaining away a weak result.
- Build one landing page. A clear headline, the price, three reasons to care and one button. The button asks for one real step: join the waitlist, sign up for early access, or pre-order with a refund promise. Say plainly that the product is coming.
- Buy targeted visitors. Spend €50 to €200 on the channel where your buyers already are, and aim for 100 to 400 visitors. That is enough to tell a clear yes from a clear no.
- Read the numbers, then decide. Compare the sign-up rate with your pass mark, and read what people searched for and asked. Then choose: go, rework the offer and test again, or stop.
What to measure
- Visitors
- People who reached the page from your ads.
- Click-through rate
- The share of people who saw an ad and clicked it. If it is low, the promise in the ad isn’t landing.
- Sign-up rate
- Sign-ups divided by visitors. This is the main number, the one you compare with the pass mark.
- Cost per sign-up
- Ad spend divided by sign-ups. It hints at what a customer may cost you later.
- Pre-orders
- The strongest signal, because people reach for a card. Always refundable, and the page says so.
Choosing a pass mark
There is no universal number. These are the starting points we use for visitors from cold ads, then adjust for the price and the market.
| The step you ask for | Our starting pass mark |
|---|---|
| Join a free waitlist | 8–10% of visitors |
| Early access with an email | 5–8% |
| Book a call or a demo (business buyers) | 2–5% |
| Refundable pre-order or deposit | 1–3% |
These are our working rules of thumb, not industry statistics. A higher price deserves a lower pass mark; a free product should clear a higher one.
Picking the channel
| Channel | Best when | Watch out for |
|---|---|---|
| Google Search | People already search for a solution | High prices in crowded markets such as law or finance |
| Meta (Facebook, Instagram) | The product is visual or bought on impulse | Broad audiences: narrow them by interest and place |
| TikTok | Younger buyers, products that show well on video | You need a short video, not a picture |
| Niche, tech and hobby audiences | Communities dislike hype, so write plainly | |
| Business buyers, reached by job title | The most expensive clicks: keep it for higher prices |
How much traffic is enough?
Roughly 100 to 400 targeted visitors. With fewer, one or two sign-ups swing the rate too much. With more, you pay to confirm what you already know. €100 usually gets you there, except in costly markets: then raise the budget or start with a cheaper channel.
A quick check while the test runs: if 200 visitors bring fewer than 4 sign-ups, under 2%, the offer isn’t landing. Change the offer before you add budget.
Mistakes that spoil a test
- Sending the page to friends and counting their sign-ups.
- Leaving the price off the page. Interest without a price tells you little.
- Changing the page halfway through the test.
- Stopping after one day. The first day is rarely typical.
- Counting likes and comments instead of sign-ups.
- Setting the pass mark after you have seen the numbers.
Keep it honest
A test page never takes money for something that doesn’t exist. Say that the product is coming. If you take pre-orders, make them refundable and say so on the page. Use people’s emails only as the page promised. Honest tests also give better data: the people who sign up for a clearly described future product are the ones you want.
When not to test this way
- When the value only shows in use, such as how a game feels. Test the promise instead, or build a small prototype.
- When the ad platforms don’t allow the product, such as gambling or medicines.
- When you will build it whatever the numbers say.