The blogFrom idea to product

How to Test Your App or Website Idea Before You Build It

Before you pay a developer in Tunisia, find out if people will really use your idea. A cheap, step-by-step way to test demand with real customers.

7 min read

The cheapest way to test an app or website idea is to get real people to take a real step before any code is written. That step can be a payment, a pre-order, a booking, a deposit or a phone number left on purpose. In Tunisia you can do most of this with a Facebook or Instagram page, WhatsApp conversations, a small ad and some honest interviews. The goal is not to prove your idea is good. The goal is to find out quickly and cheaply if it is wrong, before you spend months and savings on development.

The short answer: sell it before you build it

Development is usually the most expensive part of a digital project. Once an app is built, changing direction means paying again for design, code and testing. Changing direction on paper costs almost nothing. This is why you should test first and build second.

A real test asks people to do something that costs them a little: money, time or a promise. Saying "nice idea" costs nothing, so it proves nothing. Paying a deposit or booking a time slot costs something, so it tells you a lot.

Think of testing as a search for bad news. If your idea has a weak point, you want to find it this month, not after six months of development. Fast and cheap answers are the whole point.

Write down what must be true for your idea to work

Every idea depends on a few beliefs. Write them down on one page. Who has the problem? How often do they have it? What do they do today instead? Would they pay for something better, or switch from what they use now?

Then turn each belief into a sentence you can check. For example: "Restaurants in Sfax will pay a monthly fee for online bookings." Or: "Parents in Tunis will book a home tutor through an app instead of asking neighbours." A clear sentence like this can be tested. A vague idea like "people need a better booking system" cannot.

Start with the riskiest assumption. For most founders this is not whether the technology is possible. Almost anything can be built. The real risk is usually whether people will pay, or change their habits, for your solution.

Talk to people the right way

Interview 10 to 20 people who really have the problem. Do not count friends and family. They love you, and they will be polite. Their encouragement feels good but it is not evidence.

Ask about the past, not the future. Do not ask "Would you use an app that does this?" Ask "The last time this happened, what did you do? How long did it take? Did you pay anyone?" People are bad at predicting what they will do. They are good at telling you what they already did. Compliments are not evidence.

In Tunisia, the practical channels are simple. You can walk into shops or offices and ask for ten minutes. You can post in Facebook groups for the profession or the neighbourhood. You can follow up on WhatsApp, where many people are more comfortable answering. Watch for the signs of a real problem: they already spend money or time solving it, they get emotional about it, or they ask when they can try your solution.

Cheap tests that ask for a real commitment

After the interviews, move to tests where people must act. Choose one or two that fit your idea. You do not need all of them.

Adapt each test to how people buy here. Many Tunisian customers do not trust online payment or do not have a card. So you can test willingness to pay with a small deposit, a cash on delivery commitment, or a confirmed booking by phone. What matters is that the person takes a step that costs them something.

  • A simple page with one clear action. A landing page, meaning a single web page that explains your offer, or a Facebook or Instagram page. It should ask for one thing: pre-order, join the waiting list, or book a call.
  • A small paid ad to a precise audience. This shows whether strangers respond, not only people who know you. The cost depends on how narrow the audience is and how long the ad runs, so start small and set a fixed budget.
  • Do the service by hand first. Take orders on WhatsApp, manage bookings in a spreadsheet, deliver the result yourself. If people keep coming back while it is manual, the demand is real and you can automate later.
  • A clickable mockup. A set of pictures of the main screens, linked together in a design tool such as Figma, so it feels like an app. Hand it to people, say nothing, and watch where they get stuck.

How to read the results and decide

Before each test, write down what result counts as a yes. For example: "At least 15 restaurants agree to a free trial, and 5 pay a deposit." If you decide this after the test, you will be tempted to call weak results a success. Everyone does it, so protect yourself in advance.

Know which signals are weak and which are strong. Likes, views, shares and "I would use it" are weak. Payments, deposits, bookings and people asking again are strong. A post with many likes and no orders is a warning, not a win.

  • Clear demand: people commit and come back. Build a small first version.
  • Mixed signals: some interest, few commitments. Change the offer, the price or the target customer, and test again.
  • No interest: stop or change direction. You kept your money, and that is a good result.

When cheap testing is not enough

Some ideas cannot be faked by hand. Products built on heavy technology are one example. Regulated sectors, such as health or payments, are another, because you cannot simply improvise with patients' data or people's money. Marketplaces are hard too, because they need buyers and sellers at the same time.

In these cases, test the pieces you can. Check demand on each side separately. Check if people would pay and how much. Talk to the regulator or a specialist early if rules apply. Then plan the smallest possible first version: only the one feature that solves the core problem, and nothing else. No extra options, no admin dashboard beyond what you need, no second language until users ask. Testing the idea before spending is also a step Benders runs with founders, before any design or development starts.

Questions people ask

How much should I spend on testing before I decide to build?
Spend only a small part of what development would cost. Interviews and a Facebook page cost mostly time, while a small ad and a mockup add a modest amount. If a test starts to cost almost as much as building, you are probably testing too much at once.
Can someone steal my idea if I talk about it or put up a landing page?
It is possible but rare, because ideas are easy and execution is hard. The bigger risk is building something nobody wants because you kept it secret. Share the problem and the offer openly, and keep your detailed plans and numbers to yourself.
How many people need to say yes before the idea is worth building?
There is no universal number; it depends on your price and your market. A business tool may need only a few paying clients to make sense, while a cheap consumer app needs many more. Set your own target before the test, based on what would make the first version worth its cost.

Got an idea?Let’s start there.

Tell us about your project in a short intro call, and we will tell you where to start.

Or write to us at contact@benders.io