How to Validate Your Startup Idea Before Writing Code

To validate a startup idea before you build it, talk to people who have the problem about what they do today, test demand with a simple page or offer, and ask for money before you write any code. If strangers pay you to solve the problem by hand, you have evidence worth building on, and you will know what to build first.

Every Kenyan example in this guide is made up to show the method. Swap in your own customers and numbers.

What problem am I solving, and for whom?

Write one sentence that names who has the problem, what they are trying to do and why it is hard today. For example: "Matatu owners in Nairobi struggle to track daily takings from each vehicle because conductors report in cash at the end of the day." If you cannot name the people or the reason, keep working on the sentence before you do anything else.

Then rank the pain. A problem people face every week and already spend money on is a strong place to start. A problem they meet once a year and shrug off rarely supports a business.

How do I run customer interviews that give honest answers?

Start with a first round of, say, 10 to 15 conversations with people who fit your sentence. Rob Fitzpatrick's book The Mom Test is built on the idea that people are polite about your idea, so you learn more by asking about their lives and their past behaviour and leaving your idea out of it.

• Ask "When did this last happen? What did you do?" in place of "Would you use an app that does this?"
• Ask what they have already tried, what it cost and what annoyed them about it.
• Listen for workarounds. A spreadsheet, a WhatsApp group or a paid helper tells you the problem matters to them.
• Write down their exact words. You will reuse them on your landing page.
• End each conversation by asking who else you should talk to.

Be careful with friends and family. They want to support you, so their answers lean kind. Use them for introductions to people who have the problem.

If you meet people in person, a private room makes it easier to talk. We have four meeting rooms you can book at the house, and a day pass is KES 1,000.

How do I test demand with a smoke test?

A smoke test checks whether strangers take a step towards your product before it exists. Build one simple page that describes the offer as if it is ready, with a single action: join the waiting list, book a call or reserve a place. A free form tool or a basic website builder is enough, and no developer is needed.

Then send people to it from the places your customers already spend time: WhatsApp and Telegram groups you belong to, a small paid social media ad aimed at your city, or a flyer with a QR code at a market. Count how many people visit and how many take the action.

Set your bar before you start, for example 30 sign-ups from 300 visitors, and write it down. Deciding afterwards makes it too easy to call any result a success.

How do I take pre-sales before the product exists?

A payment is stronger evidence than a sign-up, so ask for money in a way you can honour:

• Pre-orders. Take payment for the first batch and deliver it by hand. Say you plan a meal delivery service for offices in Westlands: take five paid orders for next week and cook and deliver them yourself.
• Concierge service. Do the job manually for your first customers, with spreadsheets, phone calls and WhatsApp, and charge for the result.
• Paid pilots. For a business customer, offer a short paid pilot with a clear outcome and price, and put it in a simple written agreement.

You can collect money without building a checkout. Safaricom says Business Till applications in the individual category are processed within 10 minutes (Safaricom Business Till). Paystack's starter business tier lets you accept payments with a national ID and a personal bank or mobile money account before you register a company, up to a KES 600,000 collection limit in Kenya (Paystack Starter Business).

Be honest with early customers about what they are buying and when they will get it, and refund anyone you cannot serve. Your first customers are also your best source of introductions, so treat them well.

How do I decide whether to build?

Put your evidence side by side: what people said in interviews, how the smoke test did against the bar you set, and how many people paid. Then choose one of three paths:

• Build the smallest version that serves the people who paid.
• Change the offer if the problem is clear but your solution did not land with the people who have it.
• Stop if nobody had the problem badly enough to pay, and keep your notes for the next idea.

When you do build, build for the customers you already have. The first version can often run on WhatsApp, a form and a payment link while you learn which parts to automate.

If you want to build alongside other idea-stage founders, the Spark track of the Hackhouse Residency costs KES 25,000, runs for 12 weeks and takes no equity. HH Gen 27.1 applications are open.

Key facts

• Write the problem as one sentence: who has it, what they are trying to do and why it is hard.
• Ask about past behaviour in interviews, and leave your idea out until the end.
• Set a pass mark for your smoke test before you run it.
• A payment is stronger evidence than a sign-up. Pre-orders, concierge service and paid pilots all work before code.
• Paystack starter businesses in Kenya can collect up to KES 600,000 before registering.
• Safaricom processes individual Business Till applications within 10 minutes.

Frequently asked questions

How many customer interviews do I need?

There is no fixed number. Start with a batch, for example 10 to 15, and keep going until you stop hearing new things.

Is a waiting list enough to validate an idea?

A waiting list shows interest. Ask some of the people on it to pay a deposit or pre-order. The share who pay tells you far more than the length of the list.

Do I need a registered company to take pre-orders?

Not to start. You can take payments on a Business Till under the individual category, or on a Paystack starter account up to its KES 600,000 limit. Register your business as sales grow.

Can I validate a business-to-business idea the same way?

Yes. Interview the people who would buy and use the product, then ask for a paid pilot or a signed letter of intent. A purchase order from one business customer is strong evidence.

Sources

• The Mom Test, Rob Fitzpatrick, checked 5 Nov 2026
• M-PESA Business Till, Safaricom, checked 5 Nov 2026
• What is a Paystack Starter Business, Paystack, checked 5 Nov 2026

Go deeper with us. Join the Hackhouse community for conversations that go beyond the surface, where builders share the hard-won lessons that never make it into press releases.