In some cases, it takes no money to start a software business because the first customer (someone already in the developer's network) pays for the first product. In other cases, it takes a lot of funds before a business breaks even. Google is a good example of that. We ran a Software University Seminar that addressed this subject on November 15. You can learn more about this and order the CD by going to http://www.softwareceo.com/events/111507.php.

Sorry, but Bert's answer was pretty good considering you gave a one-line, open-ended question. The possible answers to your question are practically infinite. How small? You can start with one person calling on the phone and using email to get the first client, and then doing the work himself. That requires literally no startup capital. Is that your idea of small?

On the other hand, you could open an office in the US with 2 sales reps (one on each coast) and and an assistant, and hire 10 programmers in India. Assume some money for marketing, and some more for equipment and office space. Is that your idea of small? It isn't mine, but it may be to some people. This would take considerably more capital than the scenario above, but might be small in the opinion of some.

Im not trying to be mean--we are here to help. But if you want to get answers out of this forum that are useful, you need to ask questions with a little more specificity. The moderators aren't mind readers.

Phil Morettini
PJM Consulting
Moretti on Management Blog
http://twitter.com/TechnologyGuy
+1 858 792 1062

It should easier for you to gather the costs for the development office in India than for us. Figure very roughly 100-250K/year/head fully loaded costs for the US personnel. Another 5K/month for office and equipment costs.

Is 10-20 people enough to make a difference? That's the most difficult question of all. In my opinion, trying to offer software services in the US as an outsourcer may be the most competitive technology business I've ever seen. If you have no differentiating aspects to you company, it's not a business that I would want to enter. If you choose to do so, especially as a small startup, in my opinion it would be imperative to focus on a very tight niche--with some advantages in that niche going in (strong development experience and contacts, for example). Otherwise, it's just a price war and credibility war--and at least initially, you'd be in no position to win the credibility aspect.

Phil Morettini
PJM Consulting
Moretti on Management Blog
http://twitter.com/TechnologyGuy
+1 858 792 1062

There's no substitute for getting out the old Excel and plugging the numbers in and playing with them. A lot depends not just on the number of people but how soon the revenue ramps up. If I wanted to start a 10-person company to do PC tech support, I would not need much capital because we should be generating revenue almost immediately. If I wanted to hire 10 engineers to create a search algorithm better than Google's I might need to figure on 5 years before money started coming in.

I started my company with no capital at all. I was working for a small development company and the owner had something of a nervous breakdown. I quit and called two customers that he was not servicing (not returning their phone calls) and said I was available for contracts. I was cash flow positive from about day two. I bootstrapped that into a bunch of employees and a packaged product.