* Be OK being idle sometimes. If you double your prices and get 2/3 the work, you're way ahead. And that time isn't wasted; use it to network, to have fun side projects, to learn new stuff, to relax on the beach.
* Take bigger jobs so you can spend less time hunting.
* Prefer referenceable clients to a big portfolio. If a prospective client can call a past client and get a rave review, that's gold. Unfortunately, that means you may have to be more picky about who you take on, which works against your price bottom line.
* Move up the value chain as hard as you can. Learn everything you need to do that. Your goal is to move from "what do I code next, boss?", to being that coder, but also an architect, a project manager, a trusted advisor, an expert on product building, a UX designer, etc. Assuming you're junior, that stuff takes time to learn, so don't fake-it-til-you-make-it and it's OK to start off as a coder-for-hire. My point is just that's where you want to expand to, as opposed to adding incrementally to your portfolio. If you're senior, well, that's what makes you awesome, so position yourself accordingly.
* Other, more senior, contractors are really good lead-gen. Work with them on one project and prove your worth, and they'll fill your pipeline for you, either by passing along work they can't take on or by bringing you into larger jobs. Your network isn't just your clients.
* If it's possible, consider moving to somewhere more target-rich. Edit: now that I've looked you up, I see you're also in Cambridge; they're may be better places to find work, but I don't know them. I should disclaim, though, that I know next to nothing about the games business.
But if you are starting at $20/h or so, you might want to start at higher rate from Day 1. Because it might be easy to find clients, but often they will be a bit problematic, you work probably won't be valued greatly, etc. As paradoxal as it might seem, the best and easiest to work with clients are the ones who you are charging the most.
Related amazing podcast http://www.kalzumeus.com/2012/09/21/ramit-sethi-and-patrick-...
And the other problem is that there is lag time between two projects, so even when I have a project end, it's almost impossible to time it so that there isn't downtime in between.
As a result, I probably end up billing for maybe 20 hours a week (on average). So if I'm billing out at a rate that's twice what I would make per hour at a full-time job (40 hours a week), I end up quite a bit behind. (I have to pay for self-employment taxes, any expenses such as office space etc..., plus any benefits that I give myself). In the end, I'm learning that it's a lot harder to make the economics work out than you would expect.
When I had a startup, I looked at the guy who was billing us $140 an hour, and thought he was making bank. In hindsight, I've realized that he was probably doing just fine, but he probably did it because he enjoyed the work and the lifestyle, and not because it was going to make him rich.
Same here. Also, you can't really do more than ~5hrs/d of productive work. Personally, I don't see much point in trying to pull extra few hours when I can do the same amount of work tomorrow in 1 hour or less. It hurts the income a bit but I am much much happier.