Most first-time founders start in the wrong place. They have an idea, they get excited, and within a few weeks they are building: a logo, a website, a prototype, sometimes a registered company. Months later they find out whether anyone wants it.
Validation turns that order around. Before you spend six months building, spend 30 days finding out whether the problem is real, whether the people who have it will talk to you, and whether any of them will commit to a solution.
This is the plan we use with early founders at First500days. It takes four weeks and very little money, and you can do it alongside a job or a degree. It won’t prove that your business will succeed, because nothing can do that. What it does is stop you building something nobody asked for, and it gives you evidence you can use with co-founders, advisers and investors.
The plan has one rule: every week ends with something written down. A problem statement, interview notes, test results, a decision. If you can’t write it down, you haven’t learned it yet.
Week 1: Define the problem and who has it
In week one you write down the problem you think exists, the people you think have it, and how you’ll find them. That’s all. Don’t think about solutions yet.
Write a one-sentence problem statement
Use this structure:
[Specific person] struggles to [do something] because [reason], and today they [current workaround].
For example: “International Master’s graduates in the UK struggle to turn a business idea into a credible plan because they don’t know the local market, and today they piece it together from YouTube and university workshops.”
The workaround matters most. If people already spend time or money working around a problem, the problem is real. If they don’t do anything about it, it may be a minor annoyance rather than something they’ll pay to fix.
Narrow the customer until it feels uncomfortable
“Small businesses” is not a customer. “Independent cafés in Manchester with one to three staff” is. A narrow customer is easier to find, easier to interview, and easier to sell to. You can widen later, once you’ve won the first segment.
Write down:
- Who they are: role, situation, location
- When the problem shows up: the moment or trigger that causes it
- What it costs them: time, money, stress, missed opportunity
- What they do today: tools, people, spreadsheets, or nothing
List your riskiest assumptions
Every idea rests on a few beliefs that, if wrong, kill it. Typical ones:
- This group has the problem often enough to care.
- They see it as a problem, not just a fact of life.
- Current solutions leave a real gap.
- They would pay, or switch, to fix it.
- You can reach them affordably.
Rank these by risk. The rest of the month exists to test the top three.
Build a list of 30 people to contact
You’ll need ten conversations in week two. Reply rates for cold outreach are low, so start with about three times as many names as you need. Use LinkedIn, alumni networks, WhatsApp and Slack communities, industry groups, and introductions from friends.
End-of-week output: a problem statement, a customer profile, three ranked assumptions, and a list of 30 names with how you’ll reach each one.
Week 2: Ten customer conversations
In week two you talk to ten people who match your customer profile. You’re not pitching. You’re trying to learn how they experience the problem today.
Ask about the past, not the future
The most common mistake is asking people what they would do. “Would you use an app that…?” gets a polite yes almost every time, and it tells you nothing. People are bad at predicting what they’ll do and good at remembering what they did.
Ask instead:
- “Tell me about the last time this happened.”
- “What did you do about it?”
- “What was the hardest part?”
- “Have you tried to fix it? What did you try, and why did it fall short?”
- “Have you spent money on this? How much?”
- “Who else is involved when this comes up?”
- “Is there anyone else I should speak to?”
The last question keeps your pipeline full. Introductions from people you’ve just interviewed usually reply faster than cold messages.
Run the conversation well
- Keep it to 20–30 minutes. Ask for a short call, not a meeting.
- Talk less than a third of the time. Silence is useful, so let them fill it.
- Don’t mention your idea until the end, if at all. Once you describe your solution, people start being kind to you, and the data gets worse.
- Take notes straight away. Write down exact phrases. The words customers use will become your marketing copy.
Score what you hear
After each call, record:
| Signal | What to look for |
| Frequency | Happens weekly or monthly, not once a year |
| Intensity | They show frustration or describe a real cost |
| Current spend | They already pay for a workaround in time or money |
| Active search | They’ve looked for a better option recently |
| Pull | They ask what you’re working on, or want to be kept informed |
Compliments don’t count as evidence. “That’s a great idea” is not a signal. “Can I see it when it’s ready?” is a weak signal. “Can I pay for it now?” is a strong one.
End-of-week output: ten sets of notes, a simple scorecard, and an honest summary. Which assumptions held up, which broke, and what surprised you? If fewer than about half of the people describe the problem as painful without prompting, write that down. It matters.
Week 3: Test an offer
Conversations show you whether the problem is real. An offer shows you whether people will act on it. In week three you put something concrete in front of people and ask them to commit to it: their email, their time, or their money.
You don’t need a product for this. You need a clear promise and a way for people to say yes.
Option A: A landing page
This works well when your customers are spread out and easier to reach online.
One page, built in an afternoon with any no-code tool, containing:
- A headline that states the problem in your customers’ own words from week two
- Three lines on what you’ll do about it
- A price or price range, even if it’s only indicative
- One call to action: join the waitlist, book a call, or reserve a place
Then send people to it: your interviewees, the rest of your list of 30, relevant communities, and a small, capped ad spend if it suits your market. Track visits, sign-ups and replies.
Option B: A pre-sale
This works better when you sell to businesses or to a small, reachable group.
Go back to the people who showed the most pull in week two. Describe what you’ll deliver, when, and for how much, then ask for a commitment: a deposit, a signed letter of intent, or a paid pilot. Make it clear what happens if you don’t deliver, and offer a refund.
A pre-sale is harder to ask for than a sign-up, which is why it’s worth more. One paying customer tells you more than a hundred waitlist emails.
Set your threshold before you start
Decide in advance what result counts as a pass. For example: “At least 3 of the 10 people I interviewed agree to a paid pilot,” or “At least 1 in 10 visitors joins the waitlist.” These figures are illustrative. Pick numbers that make sense for your price and your market.
Writing the threshold down first matters. Without it, any result can be explained away as “promising”.
End-of-week output: the offer you tested, where you showed it, the numbers, and whether you hit your threshold.
Week 4: Decide go, change or stop
In week four you make a decision. Most founders skip this step. They collect evidence and then keep going anyway, because stopping feels like failing. It isn’t. Stopping after 30 days is much cheaper than stopping after 18 months.
Lay out everything from weeks one to three and give each riskiest assumption one of three labels: supported, weakened or unclear. Then choose one of three paths.
Go
The problem came up again and again in your interviews. People already spend time or money on it. Your offer hit its threshold, and at least some people committed something real.
Next step: build the smallest version that delivers on the promise to the people who said yes. Don’t build for the whole market yet. Build for them.
Change
The signal is mixed. Usually one part held up and another didn’t. Maybe the problem is real but your customer segment is wrong. Maybe the segment is right but they care about a different problem than the one you led with. Or both are right and the price or format of the offer missed.
Next step: change one thing (customer, problem or offer) and run weeks two and three again. Don’t change everything at once, or you won’t know what made the difference.
Stop
Few people recognised the problem without prompting. Nobody is paying for a workaround. The offer drew polite interest and no commitment.
Next step: write up what you learned, thank everyone who helped, and keep your notes. Founders often find their next, better idea in the conversations behind the one that failed.
A simple decision test
Answer these four questions honestly:
- Could I name five people who would be disappointed if this didn’t exist?
- Has anyone given me money, a deposit or a signed commitment?
- Do I know exactly who to build for first?
- Would I bet the next six months on what I’ve seen?
Four yeses point to go. Two or three suggest change. One or none is a stop, for now.
What this plan won’t do
Thirty days of validation lowers your risk. It doesn’t remove it. Some things to keep in mind:
- Ten conversations is a starting sample, not proof. It’s enough to spot strong patterns and kill weak ideas. It isn’t enough to size a market.
- Some products are hard to pre-sell. Deep tech, regulated products and marketplaces often need a different approach, such as expert interviews, pilots with partners, or testing one side of the market first.
- A pass in week three is permission to build small, not to build everything. Validation continues once you’ve launched. The questions just get more specific.
- Your legal and immigration setup is a separate track. If you’re an international founder, the order in which you validate, incorporate and apply for visas matters. Get advice that fits your situation before you commit to a structure.
The point of the 30 days
The first 500 days of a company decide what the founder is standing on. The first 30 decide whether you should take the next 470 at all.
Sequence beats speed. Define the problem, then talk to customers, then test an offer, then decide. Each step makes the next one cheaper and less risky. Founders who skip ahead usually end up coming back to these steps later, after spending much more to learn the same thing.
If you’re at this stage, pressure-test the plan before you spend the next six months building it. Book a 30-minute strategy call with the First500days team. We’ll go through your problem, your customer and your validation plan, and help you work out the right next step: go, change or stop.
