Over recent months, I’ve had the pleasure of working with a number of startups and developing technology businesses. It is exhilarating to spend time with someone passionate about their technology, and determined to see it succeed, whatever the environment; that makes it very satisfying to get involved and help drive these businesses forward.

In the process, I’ve had a lot of “over the beer” conversations about what you need to get a software business bootstrapped these days. The answer is, of course, “it depends” – but there are a few common threads. The principles apply equally to a backroom startup, an MBO or a research spin-out; the degree of formality, or initial cash requirements might be different, but the underlying message is the same.

If you’re going to take the plunge, these are the 5 things I think are most important.

1) Decide who you’re going to work with

You’re going to go through the peaks and troughs together. You need to have a strong relationship, and share a vision. But you don’t want a monoculture – you have to challenge each other and offer complementary skills. A technology business is a technology business. You need both aspects covered.

2) Get some basic infrastructure in place

You need good email and collaboration, web presence, telephone answering, mobile telephony, computers and internet access. Virtualize as much of this as you can (i.e. everything) – there’s no point in having on-premises infrastructure these days. It is expensive and has a tendency to consume maintenance  time that would be better spent on product development.

3) Get to a revenue model, quickly

Understanding your technology is important, but understanding who is going to pay you money, for what, and why is even more important. You haven’t got a business without that. But be prepared to throw away your first idea – don’t chase a market that should be there, but isn’t.

4) Plan to get product to market as soon as possible, but no sooner

Part of getting (3) right is testing your proposition in the market place. This is a very scary step for most people in technology businesses – there’s always something you can do to make the product better; there’s always some competitor who seems to have a killer feature you don’t. Have the courage of your convictions – build enough product to deliver on your value proposition and get it out there.

5) Work out how you’re going to support your customers

From the moment you talk to your first potential customer – even before you’ve signed them up – you’ve got to manage their expectations and give them a great experience. Work out how you’re going to do that for the whole life cycle of your product. Happy customers are your best resource, especially early on.

You can get to this point by spending only a few hundred pounds, working out of a home office, maybe only at evenings and weekends (the “midnight entrepreneur”).

When you’re there, and you’ve got a few customers, you can start to think about your scale-up cash requirements, and putting together a proper business plan for the angel networks or VC funding. You’ll have taken a lot of risk out of the equation if you’ve got real paying customers, product and a revenue model that you’re looking to scale in a substantial market, rather than a great idea that has nothing concrete behind it.