What would you do different if your startup was launching today?

For some strange reason, it's easier to see what you did wrong last time round than what you did right. I'm sure me and my team must have done a few things right as we ended up being acquired (to use Dharmesh's term, in a modest liquidity event), but I'm painfully aware of the things we did wrong:

1. Along the lines of Jarie's point about innovation - recognise that you are small start-up, not Microsoft or Oracle. In trying to build our first trading system product, we attempted to build our own messaging middleware and our own equivalent of WPF, complete with data binding and dynamic user interfaces.

   The fact that Microsoft produced such a thing a few years later meant that maybe we had great vision and insight, but trying to do in a team of three devs was just plain stupid. So the lesson there is avoid over-engineering.

2. Along the lines of Chris's point about focus and Jarie's point about distractions - having razor focus is massively important. In one sense, it matters less whether you're on the right track in terms of whatever area you go after than it does that you go after it with everything.

   In the seven years between incorporation and acquisition we took on tons of consulting work just to pay the bills, much of it unrelated to our product. In the early stages if you're self-funded, this is almost inevitable, but once you have a product, then you should make sure that any other work you do supports your product strategy.

3. Following on from the last point, second time around I would avoid growing staff at the same rate as income. We did this because we wanted to be able to produce more product, but it was self-defeating because we ended up taking on non-core work to support our high outgoings. Put another way, keep some money in the bank for a rainy day or two.

4. Finally (although I'm sure we made plenty of other mistakes), I would say that when it is time to employee a sales/marketing person (i.e., once your product is half-decent), then understand that they are very different animals than developers. Sales people work differently (especially in the enterprise space - they always seem to be socialising at lunch or dinner, when the devs are slaving over a hot compiler) - so a different management approach is needed, and they need to be properly supported by the rest of the business.

   By this I mean that, for example, when they come with a feature request for a specific client in order to clinch a deal, then it's good if the business has the capacity to digest and deliver that - we were always flat out working on the the next big thing that our sales guy always had to make do with what he had. (I'm not saying throw your product management process to the wind, but just try and be a bit flexible in support of the sales process).

Interesting your comment about founder health - there is something recent on this subject on Jason's blog I believe - I wonder if we are coming into a new generation of entrepreneurs where maintaining one's health is seen as a fundamental pillar of successful leadership. I certainly didn't do myself any favours first time round and have got quite some catching up to do before I turn fifty :-).