Okay, as one who lived through pretty much all of Loudcloud/Opsware, I can tell you that Ben pretty much nailed every single event of that 1999 - 2003 arc (when I left Opsware) pretty flawlessly.<p>What he glosses over a little (but not totally) was how fucking traumatic those layoffs were back in 2002. I recall him stating (and our VP of HR not believing he was saying it) at the completion of the cuts that there would be "no more layoffs while he was CEO". He just never wanted to go through anything like that again, so he cut as deep as he thought was need to avoid future rounds - one big round up front.<p>I don't know if it's relevant, but he also doesn't mention that a ton of employees were thinking of $20-$30 stock that then reverse split and went out at $6 (or $3 pre-reverse). The IPO was totally anti-climatic. We didn't even have a party.<p>When we had our "Opsware Founders Meeting" in Santa Cruz there was a very big deal made of the fact that we could go a _long_ time without getting a single new customer, and still not run out of cash (though, he always refused to answer my question of what Opsware would do if EDS canceled their contract with us, or just didn't pay us. :-).<p>Not running out of cash was a very, very big deal. Also, Opsware was a pretty big byzantine pieces of enterprise software (at the time, mostly written is Python, which was pretty forward looking for 1999-2000 - Ray Soursa had to convince people that no, perl would not be our platform), with all sorts of knobs and dials, and, they were still a public company, which meant there was a pretty good sized staffing overhead that your typical "startup" wouldn't see. It's not clear to me that there was much fat left on the bone by the time he was done chopping.<p>All in all though - the story is definitely worth reading over a few times - everything there happened pretty much exactly as he's calling it. Gem of an article, and Ben is easily one of the best technical company CEOs out there so that adds even more value to the read.
Shooting for the stars simply isn't in the best interests of the founders. That's why it's not popular. The most common exit now is a relatively small acquisition. All founders want (their first time anyway) is to walk away with FU money. Shooting for a $1.6 billion exit simply isn't worth the risk. Better to sell for $50 million to the guy that's shooting for the stars.
<i>"In fact, the market thought it was a terrible idea: Our stock promptly lost 80 percent of its value, putting our market cap at about $28 million. It’s worth pointing out that this was about $40 million less than the cash that we had in the bank."</i><p>Efficient market my ass.
All very good points, but Ben Horowitz actually completely missed the point of the Lean Start-up approach that people like Eric Ries are proposing.<p>"Lean" does not mean cheap. "Lean" means focused on progress, which is measured in validated learning about your customers.<p>Conflating "lean" with "cheap" serves no one. This article is arguing against cheap start-ups, not against lean start-ups.<p>This video includes some discussion, from Eric Ries, of how people are misunderstanding "lean" to mean "whatever they think will get the VCs to fund them": <a href="http://vimeo.com/9964506" rel="nofollow">http://vimeo.com/9964506</a><p>The whole "lean start-up" meme emerged with Eric Ries, as far as I know, so I think his definition of it is worth paying attention to.
This is a pretty good article. The ideas are things that both Paul Graham and Joel Spolsky have mentioned many times in their articles but it is nice to see these ideas backed up by some real experience.
Scanning the comments, there's a fair bit of argument about lean vs. cheap and about whether Horowitz is agreeing or disagreeing with the "lean" approach.<p>Afaik, all he's saying is "it depends". There may be situations e.g., if your competition is cash poor and you are cash rich, where the "fat startup" approach makes sense. It doesn't seem like he's fundamentally arguing for or against "lean" - just that a lean or fat stance only makes sense in specific contexts.