This is from an old but good interview with Tim Ferriss on his blog:

Avi: Do you have a generic method for hacking some advanced skill set. You seem to have hacked so many advanced topics that you must have a method to your madness!

Tim: Well, I do have a method and it’s really a series of questions more than anything else. It’s almost a Socratic process but I would say that, first and foremost, I have to have a very clear, measurable objective, whether that’s in language acquisition or in power lifting.

The common element is measurement, so you need to know when you have succeeded and how to measure progress to that success point, whether that’s a 500 pound dead lift or a 50 kilometer ultra marathon or getting to the point where you can do, let’s say, a single lap in an Olympic pool with 15 or fewer strokes. These are all real examples. The number of footfalls, meaning stride rate, per minute in endurance training and how long I can sustain that for say with a goal of 20 minutes at a time. Or a 95 percent fluency in conversational German as measured through different metrics. Again, all real examples.

So the first is measurement. I have a clear idea of what success looks like and how to measure it.

Secondly, I will look at the most common approaches, which are, oftentimes, the lowest common denominator but have some thread of efficacy. I will ask, “What if I did the opposite?” I’ll look at the established common practices, the established dogma, and ask myself what if I did the opposite.

If it’s endurance training, let’s look at Iron Man training, and the average is 20-30 hours of training per week for people in the upper quartile. What if I limited that to five or fewer hours per week? What would I have to do? How could I make this type of training work, or perhaps be more effective, if I had to focus on low volume instead of high volume? The same could be said of weight training. The same could be said of language learning.

If someone says it takes a lifetime to learn a language or it should take 10 years, what if I had to compress that into 10 weeks? I know it’s “impossible,” but what if? And if they say that vocabulary comes first because we should learn as we did when we were a child, which I completely disagree with – it’s entirely unfounded – what if you were to start with a radicals (Japanese/Chinese) or grammar instead?

So, flipping things on their heads and looking at opposites can provide some very surprising discoveries and shortcuts.

Thirdly, I look for anomalies. For any given skill, there’s going to be an archetype of someone should be successful at that skill. If it’s swimming, for example, it would be someone with the build of Michael Phelps. They would have a long wingspan, relatively tall, big hands, big feet and large lung capacity. So, if I can find someone who defies those anatomical proportions — say, someone who’s 5′ 5″, extremely heavily muscled, like 250, who is still an effective swimmer — I want to study what the anomalies practice because attributes can compensate for poor training. I want to find someone who lacks the attributes that can allow them to compensate for poor training.

Typically, you find much more refined approaches when you look at the anomalies. That’s true for any skill I have looked at, whether that’s programming or otherwise. So, let’s just take computer programming. If the common belief is that someone should start with language A, then progress to framework B and then progress to language C, if I can find someone who skipped those first two steps and is regarded as one of the best programmers in language C, I’m going to look closely at how they developed that skill set. In some cases, it correlates to their use of analogies and background from music or natural languages (for example, Derek Sivers or Chad Fowler )

Then I would say, lastly, is a set of questions related to rate of progress. So I don’t just look at the best people in the world; I look at people who have improved upon their base condition in the shortest period of time possible.

Let’s say I’m looking at muscular gain. I would certainly interview the person who’s, let’s say, 300 pounds and 7% body fat, but there’s a very good chance that I’ll learn more from the person who’s put on 50 pounds for the first time in their life in the last 12 months. So, I always try to establish the rate of progress and, when that person has plateaued at different points, for what duration. I find that exceptionally helpful also for finding non-obvious solutions to problems.