To validate a business idea in Singapore, you test demand before you spend money: talk to 15 to 20 real customers, run a small landing-page or pre-sell test, and confirm people will actually pay. If they will not, you change the idea early while it is still cheap to change.
Most young founders here do the opposite. They register a company, build a logo, sink two months into an app, then start asking people if they want it. By then the money and the time are gone. Validation flips the order. You gather proof first, then commit. This guide walks through the exact steps a founder aged 18 to 28 can run in Singapore, usually for under a hundred dollars.
What validation actually means, and what it is not
Validation is evidence that a specific group of people has a problem painful enough that they will pay to solve it. It is not your friends saying the idea sounds nice. It is not a survey where 80 percent tick "yes I would use this." People are polite. Politeness is not demand.
Real signals look like this: someone hands you their email to get on a waitlist, someone puts money down before the product exists, someone you have never met books a call because your offer solved their exact headache. Weak signals are likes, encouragement, and "let me know when it launches." Learn to tell the two apart and you will save yourself months.
Start narrow. A common mistake is trying to serve everyone in Singapore at once. Pick one tight group you can actually reach: NUS undergrads who need part-time tutors, HDB families in the west who want home-cooked meal prep, small F and B owners near your estate who hate their bookkeeping. A narrow group is easier to find, easier to talk to, and easier to sell.
Talk to customers before you build anything
The single highest-value thing you can do is have direct conversations with the people you want to serve. Aim for 15 to 20. Not a formal focus group, just honest chats where you shut up and listen.
Do not pitch your idea. If you lead with "I am building an app that does X, would you use it?" you get polite lies. Instead ask about their past behaviour, because past behaviour predicts future behaviour far better than opinions:
- "Tell me about the last time you dealt with this problem. What did you do?"
- "How much time or money did that cost you?"
- "What have you already tried to fix it? Why did that not work?"
- "If a solution existed, walk me through when you would use it."
Where do you find these people? For a student or NSF, your own network is a real advantage. Post in your hall or CCA chat, ask course mates, DM people in relevant Telegram and Reddit groups, stand outside the exact place your customers already gather. If you cannot find 15 people to even talk to you, that is itself a finding: your market may be too small or too hard to reach, and it is better to learn that now.
By conversation 10 you will hear the same complaints repeating. Those repeated phrases become the words you put on your landing page later. Real customer language sells better than anything you would write yourself.
Run a landing-page test
Once you have heard the same problem five times, build one simple web page that describes the solution and asks for an email or a click to "buy now." You are measuring whether strangers, not friends, care enough to act.
You do not need to code. Carrd, a free Google Form, a Notion page, or a Canva site all work. The page needs one clear headline in your customers' own words, three lines on what they get, and one button. Send a small amount of paid traffic to it, or post it in the two or three communities where your target group hangs out.
Watch the click and sign-up rate. If 100 people land and 20 leave an email for a waitlist, that is a strong signal. If 500 land and 3 sign up, the message or the idea is off. Change one thing, the headline or the audience, and test again. This whole loop can cost less than a cinema ticket and teaches you more than a month of guessing.
Pre-sell before you build
The strongest proof of all is money. Asking someone to pay before the thing fully exists is uncomfortable, which is exactly why it works. A waitlist email is a maybe. A paid pre-order is a yes.
You can pre-sell a service by simply offering to do the work manually for your first few customers. Want to build a meal-prep business? Cook for five paying families this week yourself before you think about a kitchen or an app. Want to run a tutoring platform? Tutor three paying students in person first. This is sometimes called a concierge test: you deliver the value by hand and charge for it, so you learn what people truly want and get paid to learn it.
If a stranger will not pay a deposit, you have learned the idea is not ready, and you have lost nothing but a conversation. If they will, you have a business with revenue before you have spent on registration or tools.
Build the smallest thing that works
Only after people have paid or clearly committed do you build, and even then you build the smallest version that solves the core problem. This is the minimum viable product. Its job is to keep learning, not to impress.
An MVP can be embarrassingly basic: a spreadsheet you update by hand, a WhatsApp group, a single web form wired to a payment link. Airbnb started as a photo page for two air mattresses. Your first version should feel almost too simple. If you are proud of how polished it is, you probably spent too long. Ship it to your paying users, watch what they actually do, and let their behaviour tell you what to build next.
A validation methods table you can run this month
Each method below tests a different question and costs almost nothing. Run them roughly in this order.
| Method | What it tests | Rough cost | Strong signal |
|---|---|---|---|
| Customer interviews | Is the problem real and painful? | Free (your time) | Same complaint repeats by chat 10 |
| Landing-page test | Do strangers care enough to act? | Under S$50 in ads | 15 to 20% leave an email |
| Pre-sell / deposit | Will people pay before it exists? | Free to set up | A stranger pays a deposit |
| Concierge (manual service) | Can you deliver value people want? | Free to low | First customers repeat or refer |
| MVP | Does the simplest build get used? | Low | Users return without a reminder |
Do not skip to the bottom row. Founders love building because it feels productive, but building the wrong thing well is still the wrong thing. The rows above the MVP are what stop you wasting months.
When your idea passes: the cheap Singapore admin
Once people are paying, you can register properly. In Singapore you do not need to incorporate a private limited company on day one. Registering as a sole proprietorship through GoBusiness is the fastest and cheapest path to start trading legally, and you can review the business structures and step-by-step registration on the ACRA website before you choose one. Keep records of what you earn, because business income is taxable and the rules are set out by IRAS.
If your validated idea starts to grow, Singapore has real support you can tap. Enterprise Singapore runs grants and programmes for local businesses, Startup SG lists founder schemes and accelerators, and university hubs like NUS Enterprise run incubators and mentoring open to students. Reach for these after you have proof, not before. Grants reward traction, not slide decks.
Validation is a skill you get better at with reps, and it maps neatly onto smaller experiments first. If you are still in school or serving NS, testing a small venture is a low-risk way to practise. See how to start a side hustle as a student in Singapore for a gentler starting point, and side hustles in Singapore that actually build skills for ideas that teach you something even if the first attempt fails.
Frequently asked questions
How many people should I talk to before I trust my results?
Aim for 15 to 20 conversations with people in your exact target group. You will usually hear the same core problem repeating by the tenth chat, which is the sign that the pattern is real and not one person's opinion. If everyone you speak to describes a different problem, your target group is probably too broad, so narrow it and talk to another batch.
Do I need to register a company in Singapore before I test an idea?
No. You can interview customers, run a landing page, and even take pre-orders informally while you validate. Once money is coming in regularly, registering a sole proprietorship through GoBusiness is cheap and quick, and you can move to a private limited company later if the business grows. Keep simple records from day one because the income is taxable.
What if my validation shows the idea does not work?
That is a win, not a failure. You spent a few weeks and under a hundred dollars to avoid pouring months and thousands into something people did not want. Look at what you did learn: the problem might be real but your solution wrong, or the audience wrong. Keep the useful half, change the rest, and test the next version.
How much money do I really need to validate an idea here?
Almost none. Interviews are free, a landing page can be free, and a small ad test to send strangers to it costs tens of dollars, not hundreds. Pre-selling and concierge delivery can even bring money in before you spend. If a validation plan needs a large budget up front, it is usually a build plan in disguise.
If you want to run these experiments with people who have done it before looking over your shoulder, that is a big part of what a mentor does. FINternship is a free six-week mentor-led programme in Singapore for people aged 18 to 28, and idea validation is exactly the kind of practical skill it drills. Learn the fundamentals in the masterclass, then apply when you are ready to test something real.