I'm ready to just say MVPs as a concept shouldn't be taught. In theory they should work fine, but I've never worked at a startup where they helped us. It's one

I'm ready to just say MVPs as a concept shouldn't be taught. In theory they should work fine, but I've never worked at a startup where they helped us. It's one of those things that's so easy to get wrong that it's dangerous and should be probably be avoided. He nails it when he says people focus on the "minimum" part too much. The last startup I worked at insisted on making the minimum version of every piece of a huge system. This resulted in a shallow, unpolished platform that did everything, but didn't do anything well. At every step, we'd say "this feature set sucks, and the UI is embarrassing. We need to flesh this out," and the response from PMs would be "this is the minimum viable product that might work". Well, it didn't. In the end, we made something nobody inside or outside the company wanted. We made the minimum, but it wasn't viable, and it wasn't a product. You might say "well, the fact that people don't employ it correctly is not a strike against the concept of MVP" but I've seen this happen over and over. I've only ever seen MVP go wrong, and I've only heard anecdotes about it working. Using it doesn't seem to change your odds of success as a startup. I think it may be one of those theoretically good ideas most people aren't ready for.