I used to herd lolcats, now I herd dogs. In other words, previously CTO at Cheezburger.com, and currently CTO at Rover.com. I am also a partner in Do206.com. I live online at http://scottporad.com
Wow...what a big question...here's a few answers...
The most important thing to know is that 90% of engineering management is culture management. If you build the right culture, then the engineers will manage themselves. So, do some web searches on engineering team culture and read those types of blog posts.
Also, read this: http://algeri-wong.com/yishan/engineering-management.html.
Additionally, you can read my blog, I've written quite a bit about engineering management (but I don't write as much as I used to): http://scottporad.com.
I follow Hacker News regularly. Personally, I spend my time on the "new" page instead of the "most popular".
I developed a deep understanding of Agile and Lean Startup methodologies. The key wasto understand them philosophically and strategically, as opposed to simply understanding them tactically. Once I understood them at their essential levels, the principles could be applied to all types of organizations.
Another thing I've done is to identify other organizations that served as "role models" for my team. Then, I would follow their blogs, find the slides from when they speak at conferences, etc. I would look for companies with similar technology, that appeared to be leaders in their space.
Of course, find yourself mentors with whom you can discuss your issues. Obviously, you can find those people here on Clarity.
Finally, in Seattle, where I live, we have a Seattle CTO mailing list consisting of senior engineering managers from across a broad set of companies. Find the version in your city, and if doesn't exist, start one.
After having many live demos go awry, I attempt to avoid live demos as much as possible. That being said, a few thoughts, in no particular order.
I prefer to rely on screenshots and animation which can typically illustrate my product as well, if not better than an actual demo.
Screenshots and animation allow me to craft a narrative that focuses on the products key value proposition.
A live demo often results in an audience member asking a question, then diving deep down into some sort of non-critical feature. A demo based on screenshots with animation will allow you to stay focused on our key narratives and primary value proposition.
If you must give a live demo, you could use a wireless hotspot instead of relying on somebody else's network. Another option would be to run the demo on a remote machine, and show it using a screen sharing tool (such as join.me, WebEx, etc.), so that you're only relying on the firewall to allow HTTP traffic. (Though, that might have a lot of latency.)
I am currently the CTO of a vibrant online marketplace.
As prior responders said, we started with demand. Once we had people who wanted to pay for our services, it was easier to find suppliers to fulfill the demand.
In addition, we started with a narrow focus, and then expanded. We are a local service marketplace, so that meant starting with one geography, then expanding to a second, and a third. After we had figured out the formula in three geographies, understood the dynamics of the business thoroughly, then we expanded nationwide.
As we roll out new features and services, we repeat this same pattern again and again: start in one geography, figure out the feature functionality/business dynamics/etc. then expand.
Coming back to the demand point for a moment, in every business, marketplace or not, demand is the hard problem. The fact that you're an online marketplace is irrelevant. So, to be a successful business, say focused on finding repeatable, scalable and affordable ways to generate demand and you'll be on the right path.