3 ms·
I find the "lean startup" concept to be humorous. it's like pg's y-combinator essays in MBA-ese. Build an MVP and get users to provide feedback, and iterate lik
by lmg643 13y ago
I find the "lean startup" concept to be humorous. it's like pg's y-combinator essays in MBA-ese. Build an MVP and get users to provide feedback, and iterate like crazy. As long as you're starting from MVP, the rest adheres to the principal of "JFDI". if JFDI is unclear to you, and you're looking at a flowchart to understand what JFDI means in context, then you are probably not going to understand it, ever.
- dpritchett 13y agoPlenty of people will spend years developing a product with no customers. I see small-l lean as more of a scientific method - fail fast, iterate rapidly. I'm a big fan of John Boyd's OODA loop so maybe I'm projecting.
- delinka 13y agoThe part I find humorous is constantly redefining terms. I thought "lean" referred to budgets. In that light, I would have refuted the headline's premise with a reminder that the 'outer' company was footing the bill for salaries. No, what they're discussing is yet another methodology (or an existing one with a tweak, requiring a new name...) for creating a product, getting it to [a subset of] customers, and refining it.
- mattmanser 13y agoNo, the 'lean startup' has never meant a small budget and while I love the concept, it is a pretty terrible name as it's not really cheap to do it properly. I made the same mistake until I started reading about what they were actually talking about.
- reverius42 13y agoLean process, not lean budget.
- VladRussian2 13y ago>Lean process, we've had it (and still have, just the buzzword itself somehow magically stopped flying around). It isn't lean. Process by definition can't be lean, because process itself is the fat, and once the word "process" is uttered, you may be sure that you aren't in Kansas anymore.
- nairteashop 13y ago"JFDI" is quite difficult to do in larger companies - unfortunately even most 200 person companies - when you're dealing with several stakeholders, potential bruised egos and such ("why was I not included in the startup?" "how can you launch a new product without the product team?", etc). I think the article was attempting to address that specific scenario, and not about running a "lean startup" in general.