To start a startup business, test your idea with the people you expect to pay before you spend serious money. Then write a one-page plan, register the company, build the smallest version that delivers your core promise, and launch it to a small group. Improve from what they do, not from what they say they’d do.
Most startup ideas never launch, and many that launch were never tested. This guide gives you a sequence that keeps risk low at every step, with notes for founders in Nigeria, the US and the UK.
Key takeaways
- Validate with behavior: pre-orders, deposits, sign-ups with a real cost, or pilots.
- A one-page plan beats a 50-page document. Focus it on your riskiest assumptions.
- Build a minimum viable product that tests the core promise, nothing more.
- Launch small and early. Your first users are a learning group, not a press event.
Startup or small business: which are you building?
Both are valid, and the steps overlap, but the goals differ:
- A small business serves a known market in a known way: a bakery, an accounting firm, a salon. The aim is a profitable, stable business.
- A startup is searching for a new, repeatable way to serve a market, often with technology, and aims to grow fast. It usually carries more uncertainty.
If you’re not sure, ask: “Does my idea depend on something new working?” If yes, treat it as a startup and test that new thing first.
Why most early businesses struggle
Nigeria’s national MSME survey shows how many small businesses run without the basics that help them survive and raise money.
- 41,543,028 enterprises counted in Nigeria's 2017 national MSME survey, 99.8% of them micro businesses NBS and SMEDAN, MSME National Survey 2017
- Under $300 typical total cost to register a business in the US, depending on state and structure US Small Business Administration, 2026
- £100 to register a UK limited company online, usually within 24 hours GOV.UK, 2026
The survey report itself links low business planning and low formalization to failure and to investors’ reluctance to provide capital. You don’t need a thick plan. You need a short one that you actually use, and a registered business that can sign contracts and take payments.
How to start a startup business, step by step
From idea to first users
-
Write the problem in one sentence
Who has it, how often, and what it costs them. If you can't, you have a solution looking for a problem.
-
Talk to 20 potential customers
Ask about the last time they had the problem and what they did. Listen for workarounds and money spent.
-
Run a validation test with a real cost
A pre-order, deposit, waitlist with a paid reservation, or a paid pilot. Clicks are cheap; payment is evidence.
-
Write a one-page plan
Customer, problem, solution, how you'll reach customers, how you'll make money, and your three riskiest assumptions.
-
Register the company
So you can open a business account, sign contracts and protect your name.
-
Build the minimum viable product
The smallest thing that delivers the core promise. Manual work behind the scenes is fine at first.
-
Launch to a small group
Your validation list, early adopters and friendly customers. Watch what they use.
-
Measure and decide
Keep, change or drop features based on behavior. Repeat until people come back without being asked.
For step 4, a quick SWOT analysis can sit on the back of the one-page plan. For step 6, our glossary explains what a minimum viable product should and shouldn’t include, and our free MVP scoping guide walks through cutting a feature list.
Validation test: a worked example
Say you want to launch a meal-prep delivery service for busy professionals. Illustrative numbers:
- You set up a simple page offering a first week of meals at 30 (in your currency), with a 10 deposit to reserve.
- You share it with 400 people in your target group through WhatsApp groups and LinkedIn.
- 60 visit the page, and 18 pay a deposit. That’s 18 ÷ 400 = 4.5% of people reached, and 18 ÷ 60 = 30% of visitors.
- Deposits collected: 18 × 10 = 180. If 18 paying customers can be served profitably at 30 each, you’ve proven there’s demand worth testing further.
If only two people paid, you learned that cheaply. Change the offer, audience or price and run it again. This is the core of bootstrapping: let customers fund the learning.
Registering a startup in Nigeria, the US or the UK
| Nigeria | United States | United Kingdom | |
|---|---|---|---|
| Common structure | Private company limited by shares (CAC) | LLC, or a C-corp if raising venture capital | Private limited company |
| Cost | Check current CAC fees on the portal | Usually under $300 total, varies by state | £100 online, usually registered within 24 hours |
| Useful extra | Nigeria Startup Act label for eligible tech startups under 10 years old | Free EIN from the IRS | VAT registration once taxable turnover passes £90,000 |
Sources: SBA, IRS, GOV.UK, HMRC VAT, Nigeria Startup Act 2022.
How much does it cost to start a startup?
| Cost item | Estimate | Basis |
|---|---|---|
| Company registration | Under $300 (US); £100 (UK); CAC fee (Nigeria) | SBA, GOV.UK |
| Validation tests (landing page, small ad test) | $0–$300 | Planning estimate |
| MVP | $0 (manual or no-code) to tens of thousands (custom build) | Planning estimate; scope drives cost |
| Legal basics (founder agreement, terms, privacy policy) | $300–$2,000 | Planning estimate; get quotes |
| Runway for you to live on | Your monthly costs × months to first revenue | Your own budget |
The SBA’s startup cost guide suggests counting at least one year of monthly expenses when you plan (SBA). Founders who skip this step often have to stop just before the idea starts working.
The startups I worry about are the ones that build for six months before showing anyone. A one-page site that explains the problem, plus a small ad test or twenty honest conversations, tells you whether people care, and it costs far less than a build. If nobody joins the waitlist, that isn't failure. It's the cheapest lesson you'll get.
Who you need on the founding team
You don’t need a full team to start, but someone has to cover three jobs: building the product, selling it, and running the money. Many solo founders cover one or two and borrow the rest:
- Building: a technical co-founder, a trusted development partner, or no-code tools for the first version.
- Selling: usually the founder. Nobody can explain the vision to early customers better than you.
- Money and admin: a part-time bookkeeper or accountant from day one, so taxes and records never pile up.
Write down who owns each job, even if it’s “me, for now.” Gaps you can see are gaps you can plan for.
Common mistakes first-time founders make
- Building in secret. Months of building without customer contact is the most expensive way to learn.
- Asking friends if they like the idea. They’ll say yes. Ask strangers to pay instead.
- Over-building the MVP. If you’re not slightly embarrassed by version one, it probably took too long.
- No co-founder agreement. Write down equity, roles, vesting and what happens if someone leaves, before money or customers arrive.
- Mixing personal and business money. A separate account makes taxes, fundraising and sanity easier.
Myth vs fact
- MythYou need funding before you can start a startup.
- FactMany startups validate and launch with savings and early customer revenue. Evidence of demand also makes raising money easier later.
- MythA detailed business plan proves your idea will work.
- FactOnly customers prove it. A short plan helps you pick what to test; the tests give the answers.
- MythLaunching means a big public event.
- FactAn early launch to a small group gives faster, cheaper lessons. The big announcement can come once people keep using it.
Ready to launch?
- Problem written in one sentence
- 20 customer conversations documented
- At least one validation test with real payment or commitment
- One-page plan with three riskiest assumptions
- Company registered and business bank account open
- Co-founder agreement signed, if you have co-founders
- MVP that delivers the core promise
- List of first users to invite
Frequently asked questions
What is the first step in starting a startup business?
Define the problem and test it with the people who have it. Talk to potential customers about the last time they faced the problem, then run a small test that asks them to commit money or time before you build anything big.
How do I validate a startup idea without money?
Interview potential customers, then offer a pre-order, deposit or paid pilot through a simple page or direct message. You can deliver the first version manually. If people pay, you have evidence; if not, you've saved months.
How long does it take to launch a startup?
It depends on the idea and what the first version needs. Many founders can test demand within weeks and launch a basic version to a small group soon after. Reaching steady growth takes much longer.
Do I need a co-founder to start a startup?
No, but it helps if your co-founder covers skills you lack, such as technical building or sales. If you go solo, plan how you'll access those skills through freelancers, partners or advisors.
What's the difference between a startup and a small business?
A small business serves a known market in a proven way and aims for steady profit. A startup tests a new way of solving a problem and aims to grow fast, so it carries more uncertainty and usually needs more validation.
Next step
Write your problem sentence today and book five customer conversations this week. When you’re ready for the launch itself, read our checklist of what to set up before launching a startup.
