The gap between deciding to start a business and actually trading is where most first-time founders lose momentum. Not because the work is hard, but because it is unstructured. Faced with a hundred things that all seem equally urgent, founders either freeze or do the satisfying, visible tasks — the logo, the website, the business cards — while the tasks that actually determine survival sit untouched.
Here is the uncomfortable truth that most launch guides soft-pedal: the order in which you do things matters more than how well you do any single one. A beautiful website built in week one for a proposition you have not validated is wasted effort, and the validation you skip to build it is what would have told you whether the website was worth building. A launch is a sequence, and getting the sequence wrong is how founders spend three months busy and arrive at day ninety no closer to a paying customer.
This piece lays out a structured 90-day plan in three phases, so the right work happens in the right order, and you reach the end of it with a trading business rather than a pile of half-finished admin.
Before day one: the prerequisite
This plan assumes one thing is already done: you have validated that people will pay for what you intend to sell. If you have not, stop here and do that first, because everything in the next 90 days will be wasted if the proposition is wrong. Our guide to validating a business idea covers it.
The test is simple. Can you name ten specific people or organisations who have credibly told you they would pay for this? Not that it sounds interesting. That they would pay. If you cannot, the next ninety days should be spent validating, not launching.
Assuming that is done, here is the plan.
Phase one, days 1 to 30: foundations
The first month is about getting legally and financially real, and doing the unglamorous setup that everything else depends on. Resist the urge to jump to marketing. This phase is the foundation, and foundations are inherently boring.
Week one: structure and registration. Decide whether to be a sole trader or a limited company using our structure guide, and register. A limited company costs £100 to incorporate digitally with Companies House as of February 2026. Note that identity verification for directors has been mandatory since November 2025, so allow time for it [1]. Register for the taxes that apply to you.
Week two: banking and finances. Open a business bank account, which is a legal requirement for a limited company and strongly advisable for a sole trader. Set up bookkeeping from day one, because retrofitting it later is painful. Decide how you will invoice and get paid.
Week three: the essential legals. Terms and conditions or a services agreement. Data protection basics when handling personal data, including registering with the Information Commissioner’s Office if required. The insurance your business actually needs. Our startup legal checklist distinguishes between what is genuinely required and what is merely sensible.
Week four: the minimum viable presence. Now, and not before, build the smallest online presence that lets someone find you and understand what you do. Not a perfect website. A working one. The instinct to perfect this is exactly what derails launches, because it is satisfying but not what wins your first customer.
By the end of month one, you are a real, legal, bankable business with a basic presence. That is the entire goal of phase one. Nothing here wins customers; it makes winning them possible.
Phase two, days 31 to 60: first customers
The second month is the one that matters most and the one founders are most tempted to defer, because seeking customers means risking rejection. This is where the business becomes a business rather than a set-up exercise.
Week five: define your first offer precisely. Not your whole future product range. The single, specific thing you will sell first, to a specific person, at a specific price. Narrow beats broad at launch because a precise offer is easy to say yes to, while a broad one is easy to defer.
Week six: go to the ten. Return to the people who told you they would pay and ask them to pay. This is the moment of truth, and it is uncomfortable, which is why founders find reasons to delay it by making more preparations. Do not prepare more. Ask.
Week seven: deliver and learn. Whatever your first customers buy, deliver it, and pay close attention to anything that is harder, slower, or different than you expected. Your first delivery teaches you what the business actually is, rather than what you imagined it would be.
Week eight: refine from reality. Adjust the offer, the price and the process based on what the first deliveries taught you. This early feedback is worth more than any amount of pre-launch planning, because it is real.
By the end of month two, you should have made your first sales and learned what it’s actually like to run your business. If you have not made a sale by day sixty, that is the most important signal the plan can give you, and it means returning to validation rather than pressing on to marketing.
Phase three, days 61 to 90: build the engine
The third month is about turning first sales into a repeatable process, so that month four does not require starting from scratch.
Week nine: find where customers actually came from. Look at your first sales and identify the channel that produced them. This is real data about your market, far more reliable than any assumption, and it tells you where to concentrate.
Week ten: do more of what worked. Concentrate your efforts on the channel that drives sales rather than spreading yourself across every channel. Focus beats breadth when you have limited time and one proven route.
Week eleven: systemise the delivery. Write down how you deliver, even roughly, so that the second and third customers get a consistent experience and you are not reinventing the process each time. This is the first system, and it is where scalability begins.
Week twelve: review and plan the next ninety. Look honestly at what the first ninety days produced. What worked, what did not, what the numbers actually say. Then plan the next quarter based on evidence rather than hope, because you now have it.
By the end of month three, you have a trading business with its first customers, a sense of where they come from, and a basic delivery process. That is a genuine business, and it is a far stronger position than the polished, customer-free business that founders who sequence poorly end up with.
The mistakes founders make in the first 90 days
Perfecting the website before validating the offer. The most common time sink. Effort spent making something look good before knowing whether anyone wants it.
Doing the visible work and avoiding the sales. Admin and branding feel productive and carry no risk of rejection. Asking for money carries both, which is exactly why founders avoid it and exactly why it matters most.
Launching everything at once. A single precise first offer beats a broad range. Narrow is easier to sell and easier to deliver well.
Ignoring the day-sixty signal. No sale by day sixty is information, not a reason to try harder at marketing. It usually means the validation was weaker than it seemed.
Building systems too early. Documenting a process for a business you have not tested yet is premature. Sell first, then systemise what actually happened.
Treating the plan as rigid. The sequence matters, but the plan should bend to what you learn. A founder who makes a sale in week two should not wait until week six to make more.
The principle underneath all of this
A launch is not a to-do list where every item is equal. It is a sequence with a logic: each phase makes the next one possible. Foundations enable customers. Customers teach you what to systemise. Systems let you grow. Founders who work out of order, who market before they are legal or systemise before they have sold, do the right tasks at the wrong time and wonder why ninety days produced so little.
Do the boring foundations quickly, get to paying customers as fast as the setup allows, and build the engine only once you know what actually works. That is a business at day ninety. The alternative is a very well-presented set of intentions.
Want the week-by-week plan with the templates to execute it? The Launch & First 90 Days Toolkit gives you the checklists, templates and process for every phase above, so nothing gets missed and nothing gets done out of order. For the complete founding system from idea through launch, the Startup Operating System has everything. Or book a conversation. SGI has advised more than 2,000 businesses across 47 industries since 2014.
Get the Launch & First 90 Days Toolkit
Frequently Asked Questions
How long should it take to launch a business?
The legal and financial setup can be completed in a few weeks, but a structured launch that reaches first customers and a repeatable process realistically takes around 90 days. The timeline matters less than the sequence: doing foundations, then customers, then systems, in that order determines whether the ninety days produce a trading business.
What should I do first when starting a business?
Validate that people will pay, before anything else. Once that is done, the first practical step is to choose your structure and register, followed by setting up banking and basic finances. Resist starting with the website or branding, which feel productive but do not move the business toward its first sale.
When should I build my website?
After you are legally set up and, ideally, after you have validated the offer, which is usually around week four in a structured launch. Build the smallest version that lets someone find you and understand what you do. Perfecting it before you have a proven proposition is the most common way founders waste their first month.
What if I have not made a sale by the end of the first month?
That is normal, provided you have spent the month on foundations rather than avoiding sales. The signal that matters is no sale by day sixty, after you have actively asked the people who said they would buy. If that happens, return to validating the proposition rather than pushing harder on marketing.
Should I set up systems and processes straight away?
No. Documenting processes for a business you have not tested is premature, because you do not yet know what the real process is. Make your first sales, deliver them, learn what actually happens, and systemise that in the final phase. Systems built on real deliveries are far better than those built on assumptions.
Do I need to register before I start trading?
If you are forming a limited company, yes, you must register before trading. As a sole trader, you must register for Self Assessment with HMRC, and there are deadlines tied to when you start. Either way, sort the registration in the first phase rather than trading informally and tidying it up later.
References
- Alto Accounting, Companies House fees 2026, May 2026. https://www.alto-accounting.com/insights/companies-house-fee-increase-2026
- British Business Bank, Start Up Loans, 2026. https://www.british-business-bank.co.uk/start-your-journey/finance-finder/start-up-loan
- Office for National Statistics, Business demography, UK: 2024. https://www.ons.gov.uk/businessindustryandtrade/business/activitysizeandlocation/bulletins/businessdemography/2024
- Companies House, Register a limited company, GOV.UK. https://www.gov.uk/limited-company-formation
Related Posts
Kurt Graver is the founder and CEO of SGI Consultants, a business consultancy that has helped over 2,000 entrepreneurs establish successful startups using systematic business development methodologies. An accountant with an MBA and 25 years of commerce and consultancy experience, Kurt specialises in strategic planning, market analysis, and sustainable business growth