I recently came across a post by NFX general partner Gigi Levy-Weiss: "The New Mindset for Product-Market Fit" (see the link). The main idea is that we first need to find a problem, then create and market a product promise (landing page), and only after finding a valid interest, start building an MVP.
I'm totally okay with this approach:
1. Discover a problem
2. Talk to people by manually posting on Reddit, FB, HN, other relevant communities, making calls on Lunchclub (if you want an invite, tell me), etc. Doing things that don't scale but also that costs only your time and $0.
3. Build an MVP if the problem exists
But Gigi suggests that you:
1. Find a problem.
2. Describe your solution in words and design. (I guess it means to build a landing page)
3. Market (he means launch a real paid marketing campaign on FB and measuring CTR and landing page's CTA) your product promise until you find a promise that a big enough group of people really want.
4. Then build your MVP.
What do you think? Is it okay to find the first 20-100 people by buying ads or this should be done manually and once you understand that problem exists and you know who has it and what messages work - do marketing?
ps. I love everything NFX is doing and posting that's why I want to better understand this approach to finding PMF.