Build landing page first, they said, but is it this a way to go?

In my eyes a landing page is important and should be started early on. Landing pages will change over time anyway.

You can always have some kind of sign-up for a newsletter or a waiting list. It's important to track the landing page (e.g. mixpanel or something like this). You want to know how many people come to your page, what they look at and where they click.

Your call to action must be clearly stated. The idea is that if nobody visits your landing page, nobody will get to your MVP either. If people visit your landing page but never click the call to action, it doesn't matter if whatever is behind the CTA is working or not.

A good landing page can also be a gateway to talk to people, collect contact data, ...

I've just finished MVP [1] and I would say that last thing you want is to build product and realize that you don't know what to do with it.

I would go a bit further and suggest that you don't need landing page as well. What you should do is go and talk to your customers. Use email, arrange calls, try to get confirmation that your product is what they really want and ready to pay for.

We, as hackers, tend to fall into "build it first" approach, but at the end, to find first customers, you need to "Do Things that Don't Scale" [2], which translates into talking to your audience and better to do it before you start writing MVP.

[1]: https://nofuss.io [2]: http://paulgraham.com/ds.html

The first thing is to understand a problem deeply, where and why there is gap in solutions, and an idea how to solve that. Getting a landing page up helps you validate you hypothesis. You aren't trying to capture users at this point, rather gauge how right your hypothesis is. If it only takes a day as you say, why not do that first? It also gives you a place to quickly add a demo or link to MVP.

In the end, you want feedback on your hypothesis from potential users. A landing page is one way, and nice because it is self service. There are other ways to get feedback as well.

How do you know you are building the right MVP? How do you know you are solving a problem people care about?

It depends on the details. There is no formula that is always better.

But the reasoning behind landing page first is that it focuses on engaging users/customers because some (most?) people will avoid the risk of rejection that comes from engaging potential users. Instead of building something based on people's actual problems, they build based on imagined problems.

Which points to a third approach: build something that solves one of your problems (or a problem for someone you know).

Good luck.

As a sample of 1 I want to see either a hugely compelling idea or a working example. I don't offer up my email to just anyone. I understand your dilemma.