The problem is how do you learn from this?

Okay i have a page and i HAVE to get people here with paid traffic like adwords because no one will write about my non-existent app.

So now I get a bit of traffic but nobody clicks the buy button...

Is it because i drew the wrong crowd? Is it because they don't know me? is it because adwords conversions are usually very low so do i need to pay for more traffic?

If it actually is because my product sucks, then cool.. But what do i change?

There is no way to know this; which is why i believe the buy button is not the way to go.... at the beginning.

You can still do this before you start building, but it must be done after talking to at least a few customers.

The traditional customer interviews are always the best way to start.

This post is a bit on the extreme side of not talking to your customers. Of B2B that fail, I'd be surprised if most burn out this badly. Most get someone to use them.. just not enough.

As for the advice of getting clicks, strongly product dependent.

Imagine you are a PaaS, say Heroku. It'd be a bit hard to adequately market such a product without convincing docs and people actually playing around with it. That is.. building the product.

And even if you could get some CC conversions, it still wouldn't validate the product because so much of the stick/not stick with it will come down to user experience. Which for a pretty technical product is a bit of an unknown.

Completely agree with your point.

The problem is that the majority of first time founders are: (1) overly protective about their non-existent brand (2) afraid that they will not get any clicks (3) so attached to the idea that even if no one clicks, the design is to blame not the idea

This leads into spending far too much time not iterating on a broken realization of an idea, insisting that "with time it'll work out" or "just a little further, we're gaining momentum", when evidence shows that the company is actually at its worst :P