[nubela]

You know what, I hate (its a strong word, i know) this whole lean startup movement. Yes, a lot of point make sense, iterate whats working, throw out whats not working.

But theres a giant conundrum to it, especially for certain types of startups. Lets take the startup I am working on now as an example. I am working on a Bookmarking framework for real life. Or you can call it, delicious for worthy experiences in life. But what was delicious without users? Not just do I want to bookmark material, I want to discover them too!

I had a "MVP", and it was really minimal. I spent 2 days coming up with a landing page (ctrleff.com), 1 day to plan and create the mockups, the second to actually turn it into fruition. Result? I frontpaged iPhone and Android's subreddit, and got about 10% conversion rate, while it is by no means great, it was certainly encouraging! But if MVP is all for motivation, then yes, I agree, by all means go ahead with building a MVP. But this landing page did nothing to tell me if my startup is going to be adopted or not. There were people who liked the idea. Thats it! This is no metric for any form of success. (Not every startup can be dropbox)

In my books, an MVP is an actual product that would determine actual user adoption. So the next step, or 2 months since the launch of the landing page, I'm done with the backend, and almost done with the Android App. And the app is by no means a sketchy app, I actually spent some time making sure that it looks exactly as what it should in the mockup, that it was not jerky, scrolling was smooth. Feature set? A minimal one: that is, the creation and digestion of Checkpoints (aka bookmarks). Thats it! But I am not stopping here, then there is the iOS app. Only then, would I consider my MVP done. A MVP that I won't hate to use and would continue to use. No double standards here, I will not release a product that I personally would not even use. I can be wrong, but I want to be proven wrong. Did I mention I'm a one man armt?

TL;DR: I hate the MVP movement, just build a real product with a minimal set of features.

[stanleydrew]

Sounds like you didn't build an 'MVP' according to the article then. If you read it, it complains about the same things that you do. An 'MVP' is not merely a landing page (although it sometimes might be). It is the minimal thing that you need to build to test some set of assumptions and continue learning. If you weren't satisfied with the outcome of the test, then build something else and test again. Which is exactly what you are doing.

Seems like you and the author would agree on a lot of points, including the definition of an MVP:

"A MVP is not a minimal product, it is a strategy and process directed toward making and selling a product to customers. It is an iterative process of idea generation, prototyping, presentation, data collection, analysis and learning. One seeks to minimize the total time spent on an iteration. The process is iterated until a desirable product-market fit is obtained, or until the product is deemed to be non-viable."

[loceng]

Lean startup and MVP movements only function if you have someone in control with a clear vision and understanding of what will / needs to be created, and who can defend what the MVP features need to be. If you don't understand the metrics and why they exist then you'll have no idea why your product didn't take off - and that seems too risky to spend your time, energy, money on - if you don't the strong understanding of what must be missing.

An MVP in this sense really is just a real product with a minimal set of features. :)