I was complaining that the existing website monitoring tools were too complicated and unusable, and I could probably do better in a weekend. It took seven days (https://onlineornot.com/building-saas-in-one-week-how-built-...), but I didn't have actual paying customers until I built a few more features.
At this point, I think the idea of MVP-based development is dead - you really need to look like a proper business before folks will buy, unless you're doing something completely unique. The way around this is probably to incorporate as a business with several products FWIW.
Almost three years later, I still find things to improve on a daily basis, so I guess I was wrong about that "better in a weekend" thing.
Unlike rozenmd though, it took us about a month to create the momentum we were looking for. But that eventual inertia was able to take us to being a part of TechCrunch, and then being selected to take part in a 500 Global Accelerator Program.
We found our first users when launching on Product Hunt alongside building our MVP. That's where we got more user feedback for further iterations. The MVP model still thrives in certain areas if you look hard enough.
I hope this helps!
I'm on the team "start talking about your product early on, even if you don't have the mvp ready". Don't worry about someone stealing your idea.
If the product can solve someone's problem, it is good enough, even if it won't solve all the problems your potential customers might have.
Marketing is always different. Usually I focus on getting into Google ASAP with some semi strong external Backlinks (your own sites, Reddit in the past, hackernews, etc). To be completely honest here, I pick one or two super relevant subreddits and buy a few (like 10 or 20) upvotes to get it started.
It's important to play the SEO game right, so I always have a few 'blog' articles on my sites that talk about solving an specific issue.
The following days I search the problem I am solving and plug my site everywhere. Blogs, q&a sites, forums, Reddit comments, no matter how old. Sometimes very subtle without link just the name or so.
At this point I usually have a good idea if this product is going anywhere. If it is it's time to focus on building out the MVP according to feedback, otherwise I give it another week or two and see where it's going.