I vaguely recall that the first startup I was at made it to series D. Or maybe it was just C. Either way, it was around $200M, and we ended up going bankrupt, which I attribute to a couple different things: (1) the market that the exec’s and board were targeting didn’t really develop fully—although it was probably large enough to support our small company. Which brings me to… (2) the large customers that would have bought enough to support us didn’t want to buy from a small company—they bought from more established companies, even if our product was more aggressively priced and better in every way. Which leads to the ultimate reason: (3) early on, the exec’s/board didn’t accept a buyout from one of those more established companies.
The assets were bought by a second start-up, which I believe did get its own series D, and maybe even E. The problem was they were in the same line of business and could see that the market wasn’t expanding (sales basically flat year to year and we flirted with profitability) - so they bought a third startup that was actually in a growing market but struggling financially (partially due to mis-management, I believe). This combination was going to be the key to an IPO… except that the financial crisis happened around the time. Now we were a company where sales weren’t growing fast enough to support the burn rate—so the combined company ended up getting sold to an established company for well less than the invested value. I was lucky and happy to basically get my money out of the shares that I exercised at the second start-up.
So as you can see, there are lots of ways the company might not “exit”: greed, market doesn’t develop as planned (or dries up more quickly than anticipated), exit options become unavailable (for whatever reason), exit timing becomes sub-optimal. If each of those can be discounted with high confidence, then maybe I’d consider investing.
At the risk of pointing to the obvious, the “typical” method that has been used in the past military and space is hardware redundancy (often x3).