Starting a Startup in the UK: 10 Steps to Take Before You Launch

Starting a startup in the UK comes down to ten steps before launch. First, prove the problem, size the market and study competitors. Next, scope a small MVP and a business model. Then agree founder equity, register a limited company, set up HMRC, plan funding and put simple operations in place.

Many founders take these steps in the wrong order, building first and validating later. This guide shows how to start a startup in the UK in a sensible sequence, with a checklist, a worked example and common mistakes.

Starting a startup: the 10 steps at a glance

1. Validate the problem with real potential customers.
2. Size the market from the bottom up.
3. Research competitors and the workarounds people use today.
4. Scope the smallest MVP that tests your riskiest assumption.
5. Choose a business model you can explain in one sentence.
6. Sign a founders’ agreement and set out the equity split.
7. Register a limited company with Companies House.
8. Register for Corporation Tax and understand VAT.
9. Check SEIS and EIS eligibility, then pick a funding route.
10. Set up lean operations and tooling, but hold off on heavy automation.

Step 1: Validate the problem before the product

An idea describes a solution. A startup, by contrast, needs a problem that enough people feel strongly enough to pay to fix. So start with conversations, not code. Talk to 15 to 20 people who match your target customer. Ask how they handle the problem today, what it costs them and what they have already tried.

Listen for evidence rather than compliments. “That sounds useful” means very little. Someone who already pays for a clumsy workaround gives you a real signal. Log each interview so patterns show up.

Our guide to how to do product discovery covers interviews and assumption mapping in more depth.

Step 2: Size the market honestly

Investors will ask how big the opportunity is. However, top-down figures such as “1% of a £10bn market” persuade nobody. Instead, build the number from the bottom up. Count the customers you can realistically reach, multiply by what they would pay each year and state your assumptions openly.

Keep three layers in mind: everyone with the problem, the customers your channels can reach, and the first segment you can win within 18 months. That last number matters most early on.

Step 3: Research competitors and alternatives

Your real competition often turns out to be a spreadsheet, an email thread or a junior hire, not another startup. Therefore, list direct competitors, indirect ones and the “do nothing” option. For each, note pricing, target customer and the complaints in public reviews.

Look for the gap you can own. Perhaps incumbents ignore small firms, or their tools take months to set up. A clear gap gives your positioning and MVP scope a direction.

Step 4: Scope an MVP you can actually test

A minimum viable product exists to test your riskiest assumption as cheaply as possible. It does not need to contain every feature on the roadmap. In practice, many good MVPs start as a clickable prototype or a manual service behind a simple front end.

Write down what the MVP must prove, who will use it and how you will measure success. Then cut anything that does not serve that test. If the technical side feels uncertain, an R&D validation sprint can check feasibility and build cost before you commit budget.

Step 5: Choose a business model you can explain

Decide early how the business makes money, even if you change it later. Common models include subscription (SaaS), usage-based pricing, marketplace commission and services. Each one shapes your cash flow and sales cycle.

Test the model with simple unit economics. What does it cost to acquire a customer? How much gross margin does each customer generate? How long until they pay back that acquisition cost? Even rough numbers expose weak models early. For SaaS founders, our SaaS product consulting team often helps turn these assumptions into a product and pricing plan.

Step 6: Agree the founders’ agreement and equity split

A co-founder dispute can sink an early company faster than any competitor. So put the essentials in writing before you incorporate or raise money. A founders’ agreement, later formalised in a shareholders’ agreement, should cover:

1. Who owns what share of the company, and why.
2. Vesting, so founders earn equity over time (four years with a one-year cliff is a common convention).
3. Roles, decision rights and time commitment.
4. What happens if a founder leaves early (good leaver and bad leaver terms).
5. Assignment of intellectual property to the company, including code written before incorporation.

A startup lawyer can draft these documents, and that cost stays small next to a dispute after funding.

Step 7: Register a limited company with Companies House

Most UK startups that plan to raise investment choose a private company limited by shares. It limits personal liability and lets you issue shares. You can set up a limited company online through Companies House, usually within a day or two.

To register, you will need:

1. A company name that meets the naming rules.
2. A UK registered office address and a registered email address.
3. At least one director and at least one shareholder.
4. Details of people with significant control (PSCs).
5. A SIC code describing the business activity.
6. Articles of association (many startups adopt the model articles first).

Directors and PSCs now also need to verify their identity with Companies House. After incorporation, you file annual accounts and a confirmation statement every year.

Step 8: Set up HMRC, Corporation Tax and VAT basics

Once the company starts trading, register for Corporation Tax with HMRC within three months. Open a business bank account and keep company money separate from day one. Additionally, if you pay yourself or staff a salary, you will need to register as an employer and run payroll.

VAT works differently. You must register once your taxable turnover passes the threshold, and you can also register voluntarily before that point. Voluntary registration lets you reclaim VAT on costs. On the other hand, it adds admin and may raise prices for consumers. Check the current rules on the gov.uk VAT registration guide and speak to an accountant before deciding.

Step 9: Understand SEIS, EIS and your funding options

The Seed Enterprise Investment Scheme (SEIS) and the Enterprise Investment Scheme (EIS) give UK investors tax relief when they invest in qualifying early-stage companies. As a result, many UK angel investors expect a startup to qualify before they invest. SEIS covers the earliest rounds, while EIS suits larger and later rounds.

Qualification depends on the company’s age, size, trade and the type of shares it issues. Before raising, you can apply to HMRC for advance assurance, which gives investors comfort that the company looks likely to qualify. Read the HMRC guidance on using SEIS and take advice before issuing any shares, because mistakes here can cost investors their relief.

Funding routes to compare

1. Bootstrapping: founder savings and early revenue, with full control but slower growth.
2. Friends and family: fast but personal, so document it properly.
3. Start Up Loans: a government-backed personal loan scheme run through the British Business Bank.
4. Grants: Innovate UK and similar programmes fund research-heavy projects.
5. Angel investors: often SEIS or EIS-backed, and useful for mentoring.
6. Accelerators and pre-seed funds: capital plus structure, in exchange for equity.

Step 10: Set up operations and tooling from day one

Good operations early on mean clarity, not complexity. Pick a small set of tools: accounting software, a shared drive, a simple CRM or even a well-kept spreadsheet, and one place for tasks. Name an owner for each process, even if one founder owns most of them.

What should you not automate yet? Anything that still changes weekly: your sales process, onboarding flow or pricing. Automation locks a process in place. Until the process settles, manual work teaches you more. Our guides on why SaaS founders automate the wrong things first and business process automation for startups explain what to fix first and what to leave alone.

A startup checklist you can copy

Area	Before launch	Done when
Problem	15–20 customer interviews	Clear pain and a current workaround
Market	Bottom-up sizing	First target segment defined
Competitors	Direct, indirect and “do nothing”	A gap you can explain
MVP	One riskiest assumption	Success metric agreed
Model	Pricing and unit economics draft	Payback period estimated
Founders	Founders’ agreement signed	Vesting and IP assignment in place
Company	Companies House registration	Certificate of incorporation received
Tax	Corporation Tax registration	VAT decision recorded
Funding	SEIS/EIS check, route chosen	Advance assurance applied for if needed
Operations	Core tools and owners	Nothing automated that still changes

Worked example: two founders, one B2B idea

Take two founders planning software that helps small UK accounting practices chase client documents. First, they interview 18 practice managers. Most chase by email and track replies in spreadsheets. Several already pay for a reminder tool they dislike, which validates the problem.

Next, they size the first segment and find that no competitor integrates with common practice software. So their MVP becomes a simple reminder portal for five pilot practices. They sign a founders’ agreement with four-year vesting, incorporate a limited company and register for Corporation Tax. Finally, they apply for SEIS advance assurance before speaking to angels. Throughout, they keep onboarding manual, because every pilot practice still works slightly differently.

Common mistakes when starting a startup

1. Building before validating. Months of development cannot replace a week of customer interviews.
2. Splitting equity without vesting. A founder who leaves in month three should not keep a third of the company.
3. Leaving IP with individuals. Investors will check that the company owns its code and brand.
4. Issuing shares without checking SEIS rules. A small structural error can block tax relief for your first investors.
5. Mixing personal and company money. It creates tax headaches and weakens limited liability.
6. Automating too early. Tools bought for an unsettled process rarely survive the year.
7. Hiring developers without a clear brief. Unclear requirements lead to rework, so get a translator between the business and the engineering team.

For more support on the product and technical side, see how Severus works with startups and SaaS companies.

Severus is a business automation and operations consulting firm helping companies redesign workflows, integrate systems and implement automation and AI to reduce operating costs and manual work. Once the company is running, we help founders spot operational bottlenecks early. As a result, they can grow without increasing headcount or cementing processes that do not scale.

Frequently Asked Questions

Strategy before code. Every time.

Starting a startup and not sure what to build first? Book a discovery call with Severus