How long does it take to get good at something? 10,000 hours, or 20 hours?

Author and polymath Josh Kaufman gave an excellent talk at TED recently where he discusses the horror of trying to invest 10,000 hours in a new skill in order to get really good at it. He's just become a father to a baby girl, and (like many of us) he hasn't got 10,000 seconds - let alone 10,000 hours - to invest in learning something new. So his theory (which should cheer us all up) is that the whole 10,000 hour thing is a big exaggeration. Try doing something new for 20 hours - and you should be able to master the basics

This is good advice for all of us. Learning new stuff is scary, and we tend to be bad at it at first. When someone says - you need 10,000 hours to get really good doing something, a normal human response might be to say: "fine, I give up - I'll go to the pub instead".

Kaufman argues that the biggest barrier to learning new things is an emotional one - the fear of failure. Better to give up now, then at least we can say we didn't fail after we invested all that time and effort.

But 20 hours is a lot more do-able. At least, after 20 hours, we might not be experts at something, but we'll have passed the stage of being completely useless. And, as digital artists trying to learn, say, a new piece of software, we might get past the stage of feeling incompetent. At that point, we might be able to say to a potential client: "sure, I can do that" - and then go away and practice like mad until we actually are good at it.

In fact, lots of freelancers do just this.  You might have a sketchy knowledge of a piece of software, but you can honestly say to the client that you do know how to use it, even if you're not an expert. Just make sure you're pretty decent at using it by the time the job is due.