[billroberts]

Services-based companies are of course a well-established and perfectly respectable type of business, but a very different kind of business to a product company. I worked in a service business for many years and it has good and bad points. It's less risky, because in general you don't do any work unless you know you are going to get paid for it, whereas with a product you have to do lots of development and marketing before you really find out whether your product will sell or not.

You do get lots of variety, but on the other hand you don't get so much choice or control of what you are working on.

A successful product business can be more successful (in terms of profit) than a service business. Because if lots of people like your product you sell a lot of it, with very little marginal cost of production.

For a services business to make a lot of money, then you need to have a lot of staff. Then you need an HR department and middle managers and procedures and a dress code and before you know it you're a pointy-haired boss :-) And then making sure all those mouths remain fed is a tough task for the sales team. If you are short of work, you have to keep paying the salaries and you can lose money really quickly, whereas in good, busy times, the profits you can make are capped by the number of people and acceptable charge rates.

Of course you don't need to try to get big - you could set up a small service business and keep yourself in a pretty good lifestyle, without trying to become the next Accenture.

My choice to go for a product style company was mainly because of the creative aspects of it - I wanted to be able to decide myself what would go into the software, rather than having to develop someone else's idea of a product all the time.

[twelvedigits]

Interesting points. My issue is with your contention that as a service company, you don't decide what goes into the software. This is where a web production/service agency can learn from the creative industry.

Such a service agency should position itself as an industry expert, one that clients turn to for solutions, not products. That is, if a client -- let's say, a chain of local gyms -- comes to you and needs a website built, it shouldn't be turnkey and you should expect to contribute your thinking to the end product, not just the code that creates that.

Not code monkeys. Inform and shape the final product, because you know more than your client. They need you for a reason. They don't know the best answer, you do. That's why they hired you.

[billroberts]

You're absolutely right that that's how it ought to work. And when you have a good relationship with your clients you can sometimes get close to that. In practice it won't always work like that, but you can try to pick and choose your assignments and 'train' your clients to move in that direction.