Why validate a SaaS idea before you build it?
Because the biggest risk for new software is that nobody buys it. CB Insights looked at 431 venture-backed startups that shut down since 2023 and could pin down the reasons for 385 of them. For 43%, the product didn't meet a need customers would pay for, what the report calls poor product-market fit.
Running out of money came first, cited in 70% of cases. The report describes it as almost always the final cause of death, not the root problem. Two-thirds of the product-market fit failures were early-stage companies, the ones still building their first version.
These are investor-backed startups. The lesson holds just as well for a business owner paying for a first version out of pocket: a signed development quote tells you nothing about demand. Checking demand mostly costs time. Building costs money.
How do you interview potential customers without biasing them?
Talk to them early, and talk about their problem rather than your idea. Bpifrance Création, the business creation arm of France's public investment bank, calls meeting potential customers as soon as possible “paramount” and suggests treating your idea as a pretext to understand the market. The conversation is for learning, not selling.
Eric Migicovsky, a partner at the startup accelerator Y Combinator and the founder of Pebble, suggests five questions in his talk “How to Talk to Users”:
- What is the hardest part about doing this task?
- Tell me about the last time you ran into this problem.
- Why was it hard?
- What, if anything, have you done to try to solve it?
- What don't you love about the solutions you've already tried?
He warns against hypotheticals such as “If we built this feature, would you pay for it?”. The answer describes what someone imagines doing, not what they actually do. He also says to set compliments aside: praise for the idea tells you nothing about a purchase.
Look for numbers instead: what the problem costs, how often it comes up, and whether the person has the budget and the authority to fix it. A manager who rebuilds staff schedules by hand every week and has already paid for a tool that didn't work is a stronger signal than an enthusiastic “great idea”.
Is a landing page enough to measure interest?
It measures interest; willingness to pay comes in the next step. Bpifrance Création describes a landing page, a single web page dedicated to your offer, as a way to gauge market interest and test the offer “in real conditions”. According to the same source, it doesn't take advanced design or technical skills.
For the test to be useful, Bpifrance Création lists three ingredients:
- well-written copy that describes the problem in your customers' words;
- your value proposition, meaning what the customer gains by choosing you;
- a call to action, such as a button leading to a sign-up form.
Send the people you interviewed and your professional network to the page, then count who clicks. The more the action asks of visitors, the more the result means: leaving an email address commits someone less than booking a call. Bpifrance Création also mentions crowdtesting, where an online community rates your concept through a dedicated platform.
How do you know customers will pay before the software exists?
Ask them to pay, or to commit to paying. That's pre-selling: you sell the first version before you build it. Eric Migicovsky gives a simple version for online products: instead of asking whether a feature sounds interesting, show the upgrade flow and let users enter their card details before the feature is even built.
For business software, a pre-sale can also be a signed quote, a deposit or a letter of intent, a short document in which the customer states they plan to buy. Be upfront with these early customers about what already exists, what is still to be built and when it will be ready.
Taking money requires a legal entity that can invoice. In France, Bpifrance Création points to the micro-entrepreneur status as well suited to testing an activity, provided it fits your project, and to business incubators known as couveuses, which let you test in real conditions before registering a company. It also notes that a test helps you decide whether to go ahead, adapt the project or drop it.
What should you build once the idea is validated?
The feature your first customers committed to, and nothing more to start with. Your interviews and pre-sales tell you which one it is. The rest of the spec can wait until customers ask for it.
That's how we work when we help founders launch a SaaS. The first 30-minute meeting is about identifying what your customers actually want to buy. The first module is limited to 1 or 2 features, costs a fixed €6,000 and goes live in 4 weeks, with a working demo from week two. The Module + Launch offer, at €7,500, adds a launch page and a plan for landing your first paying customers.