I'd add one more and split "product" out from "technology." For example, youtube was simply a better product for the market than google video -- even though the back-end technology was supposedly worse.
Perhaps there is the assumption that the product can easily change once google, with its massive resources, acquires the company. Evidence though, shows that buyouts and success tend more to feature-freeze products.
Oh, one more: legal protection. Again refer to youtube and the liability of some court ruling against the legal protections that online video sites currently use.
"For example, youtube was simply a better product for the market than google video -- even though the back-end technology was supposedly worse."
I know I'm going to get hammered for saying this - but I don't think back-end code is as important as a good user interface, a well designed site, and a clear vision. Your users don't get to see the back-end code, and they don't really care how it works. And servers and bandwidth are cheap, so it doesn't matter much that your code isn't well optimised.
What does matter though is that you have a clear value proposition to you users, and that they will understand this within 5 seconds of arriving to your site. This is much more important than nice and clean back-end code.
I have a bit of experience with this - I started thinking about doing a web based project management tool around a year ago, and since I didn't know any good hackers that weren't already occupied with something else I thought I would learn how to code and just do it myself. And interestingly I found that the hard part was not the coding, but the usability and flow of the site. The coding part is basically just getting stuff in and out of a database. But the flow is really hard to get just right.
I would say Youtube is quite a different monster in itself.
It's particular popularity and name recognition was important. The backroom deal made by Google with major media studios with the purchase ensured that every other flash video site would be bled to death.
The 4th reason es muy importante. So instead of working at Google, get Larry and Sergey to respect you in the morning when they buy you.
"Buying startups also solves another problem afflicting big companies: they can't do product development. Big companies are good at extracting the value from existing products, but bad at creating new ones."
Perhaps there is the assumption that the product can easily change once google, with its massive resources, acquires the company. Evidence though, shows that buyouts and success tend more to feature-freeze products.
Oh, one more: legal protection. Again refer to youtube and the liability of some court ruling against the legal protections that online video sites currently use.