Start-up purgatory: you don't go bankrupt, but you fail to build the #1 product. You have enough cash to last years, maybe even cash-flow positive, but zero chance of becoming a high-growth company. From the entrepreneur's point of view, this can be worse than start-up hell — you're stuck, working just as hard, reaping none of the rewards, watching your best people leave.

In a boom, cash is not a meaningful competitive advantage — if every company is well-funded, being super-well-funded doesn't help you win. In a bust, having a lot of cash becomes a huge weapon you can use to put enormous pressure on underfunded competitors.

During the dot-com crash, Horowitz sold Loudcloud's entire $65M/year hosting business (their only actual business) to EDS and kept the software IP. The stock dropped 80% to a $28M market cap — $40M less than the cash in the bank — but that bet became the $1.6B Opsware sale to HP.

Running fat meant laying off zero software engineers. Larger companies like IBM could always hire and train smart people; without a lasting technology-based advantage, Loudcloud had no path to beating them regardless of how lean the P&L looked.

Spending a little or spending a lot is a means, not an end. Both lean and fat are tactics in service of two priorities: winning the market and not running out of cash before you do. Making 'running lean' the end can cost you the market.

If you can raise money in a bad market, consider raising enough to wipe out your competition. Horowitz believes Bladelogic would have won the category and gotten the $1.6B exit had Opsware not aggressively outspent them on R&D and adjacent acquisitions during the downturn.

'If you are a high-tech start-up, your value is in your intellectual property. Don't stare at your spreadsheets so long that you get confused about that.'