How to Validate a Startup Idea in 7 Days (Without Writing Code)
A practical one-week validation process we run before building anything: demand tests, pricing signals, and the kill criteria we set upfront.
By FGA Labs
Every product we have ever regretted building had one thing in common: we started building before we knew anyone wanted it. So we made validation a fixed, one-week ritual that happens before a single line of product code exists. Seven days, a small budget, and a decision at the end. Here is the exact process.
Day 1–2: Write the landing page before the product
The first artifact is not a prototype — it is a landing page that sells the finished product as if it existed. Writing it forces every fuzzy idea into concrete claims: who this is for, what it replaces, what it costs. If you cannot write a convincing page, you do not have a product idea yet, you have a technology looking for a problem. The page ends with a real call to action — an email capture at minimum, a checkout link if you are brave.
Day 3–5: Buy a small amount of real demand
Organic validation takes months. Paid validation takes days. We put a few hundred dollars behind the page — search ads against the queries a desperate customer would actually type, or a handful of posts in the communities where those customers already complain. The goal is not efficient acquisition. The goal is a clean read on whether strangers with the problem will raise their hand.
- Target the problem query, not the category — 'invoice reminder automation' beats 'accounting software'.
- Send traffic to one offer, not a menu. Options at this stage destroy the signal.
- Talk to everyone who converts. Five conversations tell you more than five hundred impressions.
Day 6–7: Judge against criteria you set on Day 1
The most important part of the week happens before it starts: writing down what a pass looks like. A conversion rate, a number of conversations, a willingness-to-pay signal. Deciding the bar upfront matters because by Day 6 you are attached. You have written copy, watched sign-ups trickle in, imagined the roadmap. Without pre-committed kill criteria, every founder finds a way to call the results promising.
Validation is not about proving the idea works. It is about giving the idea an honest chance to fail cheaply.
Most ideas fail this week, and that is the system working. The ones that pass enter development with something rare: a waiting list, tested messaging, and a founder who knows the customer's words before writing the first feature.