The only valuable insights are the ones that people have used in real life. There is almost no value in listening to the blatherings of someone who makes sweeping generalizations, speaks in vague terms, and doesn't back up their assertions with any kind of data. It's the worst kind of truthiness, and HN would be much better off avoiding this kind of opinion piece.

"Personally I don’t know anything about Nick, but I do know that $30 million is a sh*tload of money to pay for a 17-year-old as an acqui-hire."

If you don't know anything, then STFU.

If you're going to criticize something as broad as "the lean startup movement" and you want to achieve anything more than FUD, then be prepared to be careful with your arguments. Have examples. Have data. Honestly, I don't know much about this "movement" but if this article is the best criticism people can muster, then I suspect it's actually quite good.

"Lots and lots of companies built around a handful of features that matter to almost no one, least of all customers."

The whole point of the lean startup philosophy is to find out how to make a product that customers DO care about and to do it as quickly as possible so you don't lose resources, time, and burnout. This whole article doesn't seem to address the core tenets of the lean startup at all. How does selling your company prematurely have anything to do with the lean startup?

All I ever hear about is the "lean startup" model. Is there an alternative... like where I commit the next 3 years to build something awesome while I work a part-time job? (my model)

Can someone think of a cool name for that methodology? I don't want to be called "fat"...

My complaints about worshipping at the lean startup alter...

The lean model only leads you to local maxima. It can't discover any feature where the sum is greater than it's parts, or critical mass is required. I'm looking at the iphone as an example.

Critical mass is another apt analogy. Just like a fusion reaction requires a certain size to become self sustaining, some businesses require more more effort to start. Pivoting is the antithetical to getting the very largest and most valuable flywheels spinning.

Not that it doesn't have it's merits, but I think lean startups are oversold.

"After all, A/B testing, or its near cousin multi-variant testing, is a known animal these days. In practice here at Bislr, we find that about 70 percent of the things we want to test are ultimately not going to make a big difference to the business."

I stopped reading after I read this. What an outrageous statement. If 70% of the things you want to test will not make a big difference, what about the other 30%? Test them! There are probably way more untested questions waiting to be answered as well.

> I think both Nick and his company would have benefited from more time as an independent company to grow into greatness.

This is half-baked reasoning. You can't use success as a rationale for why something's broken. You'd begrudge the Summly team their millions? No one's going to listen to this. Who says Nick's not going to take his cash and build the next Dropbox? It's not like a successful exit means the end of your career. If anything, that's when it really begins.

I'm not sure if I'm reading an opinion piece or an objective piece?

- Burn out? Not having sales and traction is what decides if you go home after you "go hard". Your staff burns out when they realize your reality distortion field was just hallucination and trying to fake it until you make it vooza style. Realizing startups are more about developing sales and marketing skills to find a channel to a market is the master key. This has nothing to do with lean imho. If anything, lean makes you look in the mirror and gut check more often if you're not hiding behind a keyboard.

- No Architecture? Having an architecture discussion starts with "Have you had any relationship with a code base longer than 2-3 years?" The architectural reality is the first version will change, because it'll largely be wrong. How much you plan for this is something you learn when developing any software. It's important to be kind to your future self and using enough tooling and structure that's useful, but not enough that it slows you down.I've never skipped architecture with MVPs or lean. I use light frameworks that keep me organized. Sinatra is one example, every language has one. Most importantly, I don't build my perfect empire the first go around, as a burning effigy to my brilliance to distract myself from the fact that I might not be developing as an entrepreneur to learn the sales and marketing skills to find users, or a business model.

- Investors: The day you take money, is the day you pretty much guarantee to ultimately sell and have less and less say in your company. Someone else is telling you when you go home after you go hard if founders can't figure out to make a business model out of the startup. This has zero to do with lean, if anything being the master of your own destiny is being profitable so you can have a say on doing things your own way. Being able to have more of a say in your own destiny, lean, or no lean is more about the founders and the business, sales and marketing skills that may not exist, and whether they're doing anything to learn about it.

Disclaimer: I'm a software, project and product guy from a background in lean via implementing tech in manufacturing.

I got to spend almost 10 years learning from, implementing, and re-implementing lean in a factory perspective (adding software to the process) to really have a tiny glimpse of how lean made companies like Toyota what they are since the 60's. If we look at lean on it's own, and drop startup from it for a second, lean works.

Translating lean to the startup world will only make startups into more of a science where it's an art. Like science, it should evolve, it's not perfect, but less perfect is a lack of process. I don't think the author is advocating for this at all, but in knocking lean, I would have loved to hear the steps and processes he learnt that did work for him in his scenarios.

Applying lean to the startup world is special for startups, not for lean.

For me, lean is learning to look forward and take fewer wrong steps, and to learn quicker, but to always be learning. Not using lean is like looking back and saying out of the 100 things I did, I wish I knew to find the five or 10 I needed. Lean helps narrow it down a great deal to get you on your way, but by no means is it the instant, perfect formula.

Lean tells me to shut up. Every idea, thought, innovation, revolution I have, to shut up, and simply try it and test it to see if it's something people want, hopefully enough to pay to keep me on my path of learning.

Having a business model, users, customers and making money might not be what everyone's out to do, but if it is what you're out to do, lean is definitely one way to get there.

The "Lean Startup" is about risk mitigation, and proof of market.

If you are trying to get something going, so that it can be acquired this is a good model. "Hey Amazon would be way better if it were more social, and the products were in a Magazine layout, let's be an Amazon affiliate and mock up how it looks, and get a bunch of users.... " (Pinterest)

Sometimes (like pinterest) the feature becomes enough of a draw the Lean model becomes a company, but even if it fails to reach profitability, a company could acquire it to get the user base, and avoid looking like they just knocked off a smaller site.

Occasionally the Lean Model works because a startup starts as a hobby, and the founders just need to raise marketing dollars because they have day jobs.

But if you are a company building something with a significant amount of IP, and is truly going to be disruptive the model should be hybrid. Start Lean, then go big.

My company, http://www.stremor.com is building something much bigger than what an MVP would do. Our product is like a Siri, that actually does what SRI promised Siri would do. And does it for every device with a speaker and a mic.

That's a pretty fricking huge MVP. So we put out http://tldrstuff.com to get feed back on summarization, and http://samuru.com to get feed back on search. We even had some short lived experiments to test other parts of the technology that were public, but when we learned all that we could from them, killed them so that they didn't cost money.

After our next round we will change modes. We will leave the "Lean Startup" model and move in to a model that favors fully developed product, has a heavy marketing budget, and focuses on not just being "an interesting take" on the space, but a "Serious competitor" with "Industry leading technology". (These are in quotes because because that is the kind of stuff you see in press about MVPs and Full products respectively.)

The real challenge for a startup is knowing if the MVP is 1000 lines of code, 10k lines of code or 400k lines of code.

You can't have an MVP that competes with Facebook with out having most of Facebook's features plus some new features. But you can an an MVP that changes the way people schedule Dog Poop removal from their lawn with very little code because it is a Business model start up not an IP start up.

I guess to really sum it up, You need to know if you are an IP company, A Feature Company, An Innovative business model company, and how mature your competition is, because the maturity of the space determines how Minimal your MVP can be.

"Lean startup" is often a euphemism for poorly-thought-out designs, long hours and understaffing, poor compensation, and reckless, political firing. I'm sure that's not what Ries intended but the psychopaths ruined "lean startup", I would argue.

yawn - another article slamming the "lean startup" approach, written by somebody who clearly doesn't have a clue about the "lean startup" approach, but somehow still appears to have an axe to grind. This article is almost totally devoid of value and basically just amounts to linkbait. If you haven't clicked through and read it yet, I'd recommend moving on to something more useful.