Make a hypothesis, let the market validate it. I would say if you're spending more than 1-2 months on designing, developing, and testing (on real cusotmers) yo

Make a hypothesis, let the market validate it. I would say if you're spending more than 1-2 months on designing, developing, and testing (on real cusotmers) your MVP, your cycle is probably too long. I'm currently looking at a 6 week cycle for one "iteration". It will depend on how difficult your product is, as well. There was a linked article on HN recently about how dropbox's MVP was too complex to build ahead of time, so they used a video demo as their MVP.