When we started working on AI Notebook with the team, it was easy to jump straight into questions like:

  • What features should it have?
  • What should it look like?
  • How could it be different from the notebooks already on the market?

But our first day started with a more basic question:

  • Would anyone actually buy this?

That was why I shared a validation roadmap with the team. It had nine stages and four points where we would need to stop, review what we had learned, and make a decision.

The idea behind it was simple.

Instead of spending months building a polished product and only then finding out whether people wanted it, we would break the process into smaller steps.

After each step, we would look at the evidence and decide whether to keep going, change direction, or stop.

This matters even more when you are building hardware. We cannot make a device just because the concept sounds exciting or because the team likes it.

There are production costs, suppliers, minimum order quantities, and plenty of decisions that become difficult to undo once money has been committed.

So before getting too deep into how to build the product, we needed to answer three things:

  • Is this a real problem?
  • Who feels it most strongly?
  • Do they care enough to take some kind of action?

Our initial plan was to research the market and potential customers first, then narrow down the product concept.

If the evidence looked promising, we would create a mockup, build a landing page, and run a small test to see whether anyone was willing to join a waitlist.

At that point, the landing page would not be a launch site.

It would simply be a way to ask the market a clearer question.

Instead of asking people, “Do you like this idea?”, we wanted to see whether they would click, leave their email, or come back to learn more.

Even those signals would not prove that someone was ready to buy. But they would tell us more than a polite compliment and help us decide what to do next.

The rule we set on day one was pretty straightforward:

  • Only build when there is a signal. No signal, no build yet.

The roadmap could not guarantee that we would choose the right product. It could only help us find out sooner if we were wrong, before putting too much time and money into an assumption we had not tested.

So the first day of AI Notebook did not begin with a design or a feature list.

It began with finding a good enough reason for this product to exist.

Then we started researching. The more we read, the less certain some of our original assumptions felt.