> I don't get why people are so keen on the per-hour model. There's not going to be one correct answer to this question - per-hour rate or total project cost.

> I don't get why people are so keen on the per-hour model. There's not going to be one correct answer to this question - per-hour rate or total project cost. And regardless of which one is best an hourly rate is still a useful productivity metric. With the hourly model you're protected as a freelancer. Regardless of the number of revisions, if the project takes 10 hours or 100 hours, you will get paid. (Though if the hours required are going to increase 10-fold you'd better communicate that to the client.) I have several clients who prefer the hourly rate model. Which is fine by me. It helps me hit my weekly/monthly/quarterly goals. Also, we both save on the administrative headaches of having to spec out, estimate and approve every single change. And as a freelancer anything that reduces my non-billable work is a big plus. The downside for me: the client gains any increase in my productivity. If I increase my productivity per hour by 25% it takes me less time and I make less money. But the client still receives the same amount of value from my work, and pays less for it. And since most skilled workers become more skilled and more prolific over time this productivity transfer almost always occurs. Fixed cost projects can be more lucrative since I pocket any productivity gains. Imagine I've secured a $2000 fixed-cost project. If I can knock it out in five hours, I've secured an effective rate of $400/hr. The downside, of course, is if my productivity lags - say I spend 100 hours on the project - then my effective rate plummets to $20/hr. Every fixed-cost project should be measured hourly. Even a big-ticket project will suck if you work yourself into a minimum wage to complete it. I use the hourly metric to determine productivity. When this metric starts to drop during a fixed-cost project, that's when I reach out to the client to figure out a budget increase. So - which one is better? Hourly or fixed-cost? Most of the time it doesn't really matter. If I'm working a fixed-cost project and my productivity metric falls, I ask for more money. If I'm working an hourly project and it's going to take more time implement revisions, I'll tell the client and get approval. Regardless of the project type if the scope changes you're going to need to ask for more money/hours. BUT - if the project is in a domain I have deep knowledge in, and I can be relatively assured of a fixed scope (with only minor changes), I will go fixed-cost every time. I want to capture my productivity and maximize my profit from it.