I think the Lean methodology has some good ideas, and I often follow it myself. Someone once compared it to aspects of scientific methodology, and I agree and

I think the Lean methodology has some good ideas, and I often follow it myself. Someone once compared it to aspects of scientific methodology, and I agree and think it is something to think about. You do not have ideas, you have hypotheses. You might feel you have very good hypotheses that will pan out, but there is one great unknown which you do not know about and that is user reception. If someone like Steve Jobs had a 100% batting average at such things, then Lisa and NeXt would have been runaway hits. You are preparing to do experiments, and that experiment is what no one can predict, public response. You have not done any experiments yet, you are just modifying hypotheses before you start, which is fine in some circumstances. I can't help but think it is psychological. People don't want to put their effort into a product they think will be great and have it rejected by the public. They take the rejection personally, as if they themselves are being rejected. This goes back to the discussions on HN about how San Francisco is one of the few places where failing at a startup is not seen as a badge of shame. The normal view is that releasing a product which is rejected is a failure you should be ashamed of. This standard view you probably hold, perhaps unconsciously. The lean view that a bad launch is not a big failure to be ashamed of is not a normal view for the average person. I don't think you've taken proper account of how different the lean, Bay Area startup view of failure is different than almost everywhere else. You're acting "normally", which is what the problem is. Most people avoid failure, while you're working to increase your chances of failure. "Improvements of existing ideas" - yes an improvement, but is it an important improvement or an uninmportant improvement? Important means the public wants it, unimportant is a "nice to have". I have released minimal viable products, with my hand still containing a list of features I left out of that lean, minimal first product. I was always sure customers would demand these features first. They usually did not. They usually asked for things I would have never thought of. Sometimes their requests and suggestions were very good. Sometimes I would have three separate people request a feature I had never thought of. Yes, those "improvements" I thought of myself before releasing, I could have implemented them by postponed the first release of my minimal viable product. They would have been nice to have. But no one ever asked for those features any how. And the features everyone asks for I had not even thought of. You might be smart, but you are not smarter than the combined brainpower of dozens, or hundreds, or thousands of your customers. The bottom line is the best product your customers will want is one released minimally viable at first, and ultimately shaped by their requests. By locking yourself away and building what you think they should have, you're not giving them an awesome product on first delivery, you're cheating them of input into the process. The features you think are so necessary are not. No matter how long you take to ship version 1, it will be lacking features users want.