Validate your business idea before launch: the method

Quick answer. Validating a business idea means securing proof of paying demand before you spend money. The method has six steps: hypothesis, customer interviews, MVP, pre-sales, indicator tracking, and a go, pivot or no-go decision. Useful benchmark: per INSEE, 69% of French companies created in 2018 (excluding micro-entrepreneurs) were still active five years later.

An idea is only worth something if someone is willing to pay for the solution. Many founders reverse the natural order: they register the company, open a bank account, buy a website, and only then look for their first customers. Validation flips that sequence, and it is what protects your cash position.

The data calls for caution without drama. According to INSEE, 69% of companies created in the first half of 2018, excluding micro-entrepreneurs, were still active five years later. Survival varies sharply by legal form and sector: 70.6% for companies versus 63.3% for sole proprietorships, and only 64% in retail and wholesale trade. Testing demand upstream tilts these odds in your favour.

Across the incorporation files we support, the most common roadblocks are neither legal nor tax-related: they stem from overestimated demand. The founder mistook polite interest for purchase intent. Validation exists precisely to separate the two before you invest.

Validating is not asking your circle whether they like your concept. It is testing a precise hypothesis against real, measurable behaviour. A good hypothesis names three things: a problem, a target, a solution. As long as it does not fit in one sentence, the project is too vague to test.

There are three levels of proof, from weakest to strongest:

• Declared proof: the customer says they have the problem. Useful, but fragile.
• Behavioural proof: they spend time, sign up, leave their details. More solid.
• Monetary proof: they pay, even a deposit. This is the only genuinely reliable proof of market.

The goal of any validation effort is to reach monetary proof as fast as possible, without having built the final product. This is where the minimum viable product, or MVP, becomes meaningful.

Here is the procedure we recommend, ordered to keep spending low until demand is proven. Before starting, it often helps to run a market study in-house to frame the target.

1. Frame the problem-solution hypothesis. One sentence: this customer suffers from this problem, my solution solves it this way. If you cannot name a precise customer, start over.
2. Interview 15 to 30 target customers. Discovery interviews focused on their past (how do they solve the problem today, what does it cost them), not on your product. Beyond 20 interviews, answers repeat: a sign your target is well defined.
3. Build a minimal MVP. The smallest version that solves the problem: a landing page, a clickable mockup, or a service delivered by hand before any automation.
4. Attempt paid pre-sales. Ask for a deposit, a pre-order, a billed trial subscription. A financial commitment settles the debate better than any survey.
5. Measure conversion and acquisition cost. How many visitors become customers, how much each acquisition costs, what the average order value is. These ratios reveal whether the model scales; they foreshadow your unit economics.
6. Decide go, pivot or no-go. Compare the evidence against thresholds set in advance. The decision is binary and owned.

The strength of this sequence is financial: each step costs more than the last, but is only triggered if the previous one convinced.

The number of interviews depends on how diverse your target is. Fifteen are enough for a homogeneous niche market; thirty are useful for a segmented mass market. The right signal is not a round number, it is saturation: when a new interview adds no fresh information, you can stop.

On budget, the table below gives orders of magnitude observed during validation phases run alongside salaried work or an unemployment-benefit period.

These figures do not replace a tailored estimate. Once validation holds, we help you build a financial forecast that turns these early signals into revenue assumptions and a break-even point.

Three numbers deserve tracking as soon as pre-sales begin, because they govern future viability.

You can extend the analysis by estimating your break-even point with our tool to calculate your break-even threshold. A project that proves demand but whose acquisition cost durably exceeds the margin is not validated: it is flagged as unprofitable, which is precious information before incorporation.

Pre-sales before incorporation. You may collect pre-sales as an individual before creating the company, but Article L210-6 of the French Commercial Code makes founders jointly liable for acts carried out on behalf of a company in formation, unless the company takes them over after registration. Keep a record of these commitments so they can be cleanly assumed. On the customer side, Article L221-18 of the Consumer Code grants a 14-day withdrawal right on distance sales: build it into your terms.

Testing under the micro-entrepreneur status. Many founders validate by first invoicing as a micro-enterprise, whose 2026 ceilings reach 203,100 euros for sales and 83,600 euros for services. The VAT exemption (franchise en base) applies up to 85,000 euros (trade) and 37,500 euros (services): while you stay below, you charge no VAT, which simplifies a test phase. You will then need to choose your legal form for the longer term.

Validation funded by unemployment benefit. A jobseeker may opt for the ARCE, namely 60% of the remaining ARE rights, paid in two instalments (50% at creation, 50% six months later), after a 3% deduction. This capital can fund the test phase, but it locks in a choice: it cannot be combined with continued benefit payments. The decision deserves a cool head.

• Courtesy validation. Your relatives will almost always approve. Seek strangers within your target, not encouragement.
• The overbuilt false MVP. An MVP that already looks like the final product is no longer a test, it is a disguised investment. Cut it further.
• No decision threshold. Without a go or no-go criterion set in advance, you will always read results in your own favour.
• Spending committed too early. Website, logo, stock, premises: until demand is proven, these costs destroy cash. To gauge the stakes, see the real cost of creating a company.
• Skipping the contract framework. A pre-sale is still a sale: terms and conditions, withdrawal rights, invoicing. A legally sloppy test creates a real risk.

Recently, a director in career transition asked us to structure a consulting activity for which he had already invested in a website and a brand. On review, no sale had preceded that spending: the enthusiasm of his circle had stood in for validation. We recommended pausing the costs and first testing three paid offers with real prospects. Two triggered no order, the third converted. The project was refocused on that offer, with cash preserved.

Our reading is consistent: validating a business idea is first a financial discipline. Every euro spent before the first sale is a bet; every euro collected before launch is a certainty. That is why we treat the test phase as a mini business plan whose output is not a product but a documented decision. As chartered accountants registered with the Ordre, we stay focused on what falls within our mission: reading the indicators, framing the forecast and securing the first cash flows.

The underestimated risk is not the test that fails, it is the test that half-succeeds. A project that sells but with an acquisition cost above its margin gives the illusion of validation. Our role is to make that point objective before you commit irreversible structural costs. This often means weighing a quick launch as a micro-enterprise against a more structuring company incorporation.

Hayot Expertise tip. Set your decision thresholds before launching any test, not after. Reserve company incorporation for the moment monetary proof exists. If you are unsure how to read your first numbers or which legal form to choose, we frame the validation phase with you, from the first customer interview to the forecast.