I've been to quite a few conferences, speaking about my startup's software (http://pagekite.net http://pagekite.net). In my case, since we release our core pro

I've been to quite a few conferences, speaking about my startup's software (http://pagekite.net http://pagekite.net). In my case, since we release our core product as Free Software and are very aligned with the decentralization and "cloud backlash" movements, it's been easy to get speaking slots to get up on stage and talk. This has been an absolutely fantastic learning experience, even though it also taken a huge amount of time. Being able to get up on stage and comfortably explain to a room full of people is a very valuable skill, and one you can really only learn by doing. Even if you are just standing in a booth, you can cultivate this sort of thing by actively seeking out conversations with passers-by. Another thing, is we've also noticed there is often a 6-month lag or even more before seeing any tangible returns - people listen, file away the ideas and then come back when they are ready. So although I agree that it is important to carefully evaluate the costs involved in events like this, I think many of the metrics used in the article may be a bit short-sighted. Basically, "wasting time" is part of the learning experience. If you don't do it, you never learn.