Every ambitious founder will, at some point, face an important question.
How do you know whether you are building a business customers genuinely want—or simply becoming very busy building what you hope they will want?
Early on, it is easy to equate activity with progress. Building the brand, refining the product, adding features, and preparing to launch all feel productive. Yet genuine learning can sometimes lag behind the effort.
The Lean Startup by Eric Ries introduced a more useful way to think about early-stage growth: treat every new idea as an assumption that needs to be tested. Build the smallest useful experiment, measure what customers actually do, and use what you learn to make the next decision.
The goal is not to limit ambition, but to shorten the time between having an idea and understanding whether it is worth further investment.
The Expensive Version of Optimism
Imagine a founder who has identified a frustrating problem in a growing market. The idea receives an enthusiastic response from friends, colleagues, and potential partners.
Encouraged, the founder spends six months developing a polished platform. A designer creates the brand. A developer adds dashboards, automated reports, and several pricing options. Social media content begins before a single customer has paid.
Launch day arrives. People like the posts. They congratulate the founder. A few book demonstrations. Very few buy.
It is easy to assume the challenge is visibility and to invest more in marketing. Often, though, the real question is whether the problem is urgent, the solution is valued, or the buyer is ready to make a change.
One of the most important lessons for any founder is that praise is not proof, interest is not demand, and a finished product does not guarantee a validated business.
Before asking, “How do we launch this?”, it is worth considering:
“What is the riskiest assumption behind this idea, and what is the most practical way to test it?”
The Build–Measure–Learn Loop
The central idea in The Lean Startup is a continuous learning loop. You build an experiment, measure the response, and learn whether to continue, change direction, or stop.
Build
Create the smallest version of the idea that allows you to test the assumption. This may be a landing page, a manual service, a clickable prototype, a paid workshop, a pre-order or a pilot with one customer. It does not always need to be software.
Measure
Choose evidence that reflects real behaviour. Track whether people book, reply, introduce you to a decision-maker, agree to a pilot, use the solution repeatedly, or pay.
Avoid relying solely on impressions, compliments, follower counts, or survey responses about what people might do.
Learn
Use the evidence to make a decision. Should you persevere, adjust the offer, change the customer, solve a different problem, or abandon the idea?
Learning only creates value when it changes what happens next.
This cycle repeats, each time helping to remove a key uncertainty and bring greater clarity to the business.
Start with the Assumption Most Likely to Break the Business
It is common to test what feels easiest, such as names, colours or features. While these questions can seem productive, they often leave the most significant risks unaddressed.
A stronger approach is to write down the assumptions that must be true for the business to work:
- A specific group of customers experiences this problem often enough to care.
- The problem is important enough for them to change their current behaviour.
- The proposed solution produces an outcome they value.
- The person who feels the problem can influence or make the buying decision.
- The business can reach those customers at a sustainable cost.
- Customers will pay enough for the model to become viable.
- The team can reliably deliver the promised outcome.
Consider which assumption carries the most uncertainty and could have the greatest impact if it proves false. That is often the best place to start testing.
If customers are not willing to pay, adding more features will not change the outcome. If the buyer cannot approve the purchase, additional meetings are unlikely to create a sustainable route to market.
A Minimum Viable Product Is an Experiment, Not a Cheap Final Product
A minimum viable product is often misunderstood. It is not about releasing unfinished work or offering something unreliable. Instead, it is about creating the simplest experience needed to learn something valuable.
For a founder considering an AI recruitment tool, the first experiment may not be a complete platform. It could be a manually delivered service for three recruitment agencies: the founder reviews a live workflow, uses existing tools behind the scenes, and produces the promised result.
This tests whether the outcome matters before the company invests in automation.
For a food entrepreneur, it may be a paid tasting event or a limited-run product sold through a single retailer. For a business-to-business service, it may be a tightly defined paid pilot with agreed success measures. For a marketplace, it may start with manual matching on both sides.
The strongest early version is not the one with the most features, but the one that delivers the clearest learning with the least unnecessary investment.
Do Not Confuse What Customers Say with What They Do
Customers are often generous in conversations. They may say an idea sounds useful because they want to encourage you, because the imagined product is appealing, or because saying yes costs nothing.
Behaviour provides stronger evidence. A useful evidence ladder might look like this:
- Weak signal: “That sounds interesting.”
- Better signal: The customer agrees to a second meeting and brings a colleague.
- Stronger signal: The customer shares data, gives access to a live process, or commits staff time.
- Very strong signal: The customer signs a pilot agreement, pays a deposit, places an order, or uses the solution repeatedly.
Not every early conversation will lead to a sale. The key is to recognise the difference between encouragement and genuine commitment.
Choose Measures That Help You Make Decisions
It is possible to gather a great deal of data without gaining real insight. Website traffic, downloads and social engagement are useful only when they connect to the behaviours that matter for the business.
A better measurement plan begins with a decision. For example:
- Should we continue targeting this customer segment?
- Does this problem occur frequently enough to support a paid solution?
- Can we convert a first conversation into a paid pilot?
- Do customers reach the promised outcome?
- Do they return, renew, or recommend us?
- Can we acquire and serve them at a viable cost?
Once the decision is clear, focus on the few measures that will inform it. A metric is valuable when it guides your next step, not simply when it shows activity.
Know When to Persevere and When to Pivot
A pivot is not a failure or a complete restart. It is a deliberate course correction designed to test a new fundamental assumption while preserving what the team has already learned.
You may keep the problem, but change the customer. You may keep the customer, but focus on a more urgent problem. You may replace a subscription with a service-led model, narrow the offering to a single valuable use case, or change the route to market.
Decisions are best made using evidence gathered over a set period, rather than reacting to a single disappointing week or an encouraging conversation.
Before starting an experiment, agree on what result would encourage you to continue, what would prompt a change, and what would signal that it is time to stop. This helps keep decisions clear and consistent.
A Practical Seven-Day Validation Sprint
You do not need a large research budget to get started. In just one focused week, it is possible to replace several assumptions with practical evidence.
- Day 1: Define the decision. Write down the idea, the intended customer, and the single assumption you most need to test.
- Day 2: Speak to the market. Arrange five conversations with people who experience or buy around the problem. Ask about recent behaviour, current alternatives, cost and urgency.
- Day 3: Design the experiment. Choose the smallest credible test: a paid pilot, a pre-order, a prototype, a landing page, a manual service or an offer sent directly to relevant buyers.
- Day 4: Set the threshold. Decide in advance what evidence would count as a promising, inconclusive, or negative result.
- Day 5: Run the test. Put the offer in front of real prospective customers. Ask for an action that requires commitment, not simply an opinion.
- Day 6: Review the evidence. Separate facts from interpretation. Record what customers did, where they hesitated, and what repeatedly mattered to them.
- Day 7: Make the next decision. Continue, change one assumption, or stop. Define the next test before returning to product development.
A Founder’s Learning Test
Before investing the next month in your idea, answer these questions clearly:
- Who is the specific customer?
- What problem are they already trying to solve?
- What evidence shows the problem is urgent?
- What do they use or do today instead?
- Who makes the buying decision?
- What is the riskiest assumption in our model?
- What is the smallest test that could challenge that assumption?
- What action will we ask the customer to take?
- What result would persuade us to continue?
- What result would make us change direction?
- What have we learned in the past 30 days that altered our plan?
- Are we investing in evidence—or using activity to avoid an uncomfortable answer?
If the answers are unclear, the next step may not be further development, but rather a more focused experiment.
Build the Business and the Learning System
A successful start-up does more than create a product. It creates a reliable way to learn what customers need, what they will pay for, and what can scale.
This discipline only becomes more valuable as the business grows. Teams that test early tend to make smaller mistakes, protect cash, focus their development, improve their sales message, and provide investors with stronger evidence that the opportunity is real.
Without this evidence, it is easy to spend limited time and resources perfecting something the market may not need. With it, you can move forward with greater clarity, build around genuine demand, and show customers, partners, and investors why the opportunity matters.
Your Next Step
You do not need every answer before you start. What matters is a clear problem, a practical test, and the willingness to let customer evidence guide your next steps.
The plan is simple:
- Define the customer and the problem that matters most.
- Test the riskiest assumption with the smallest credible experiment.
- Use what customers do—not simply what they say—to decide the next move.
At The Mill Enterprise Hub in Drogheda, we help ambitious founders and businesses test ideas, clarify their proposition and connect with the people, programmes, expertise and opportunities that can support their next stage of growth.
If you have an idea to validate or a business you want to strengthen, we invite you to connect with The Mill team to discuss your next steps.
The aim is not to launch the most polished version of your first idea, but to learn quickly enough to build the right business for your market and community.
