I've done this successfully for a while but not extensively, and I've watched colleagues and friends try it with mixed results, so for what its worth I've seen

I've done this successfully for a while but not extensively, and I've watched colleagues and friends try it with mixed results, so for what its worth I've seen a few common failure modes. Roughly in order of impact. #1 - to be a consultant you need a consulting business. If you aren't working for someone else, you need to create it. The two things technical people often totally underestimate is the sales aspect (25%+ of your time), and the time spent on other non-billable work. Unless you are pretty niche and already well known, you are likely going to have to find the (right) work more than it finds you, at least at first. #2 Taking on too much and/or the wrong kind of work just because you can get it. If you establish yourself as a generalist, you're going to end up only competing with service companies and other generalists for lowest cost. Also, once an organization has hired you as "an X", it is harder for them to consider you as "a Y". If you do a basic programming contract with them because you need the work, you may never be seriously considered for anything else (by them). I probably said no to more than 1/2 the contracts I was offered when I started out, and it was a good choice even when I did have some down time. #3 charging too little. You want to be doing impactful, high value work. Or else you are probably really a freelancer looking for long term contracts and work/life balance you control (not that this is a bad thing, but it's not the same) #4 charging too much (for the work you are doing). In order to make the balance sheet work you are pricing yourself at the top of a niche and not getting enough work. Instead you should be doing something higher value. I've seen people oscillate for years on a 2 months of work, 3 months of no work, etc. cycle this way. For me this is a sign you are going after the wrong kind of work. #5 focusing on the work you enjoy, rather than the impact you can make - this is also a viable route for a lifestyle adjustment but it's only efficient if you are lucky. Some other thoughts: I think there are natural sizes for this sort of business and it has implications for the types of work you compete on. You can use sub-contractors (e.g. consultant/freelancer colleagues) to manage bandwidth a little but but there is overhead. I suspect it's a lot easier to keep 6 employees fed than 2, if you go that way. It's sort of a restatement of #1 - but network is critical.