[timruffles]

Could be right there :) Inference from either failure or success is tricky.

That said, I think "not knowing my customer" is still a sound lesson. If you know your customer, you'll know whether there is or isn't a market.

Re post-mortems, should keep you going:

http://www.chubbybrain.com/blog/startup-failure-post-mortem/

http://www.businessinsider.com/the-25-best-startup-fail-stor...

[asanwal]

Another related one which analyzes 32 post-mortems and identifies the 20 recurring reasons for failure

http://www.chubbybrain.com/blog/top-reasons-startups-fail-an...