Setting Effective Personal Goals

So you’ve decided you want to set yourself some personal goals to motivate yourself to achieve more. Great! But how to choose the right goals?

One common piece of advice, almost a cliche at this point, is to set goals that are SMART: Specific, Measurable, Achievable, Relevant and Time-bound (different sources will give you slightly different expansions of this acronym). I don’t disagree with any of this.

But how do you tell when you have picked the right level of challenge? You might constantly hit your targets, but have a nagging feeling that you aren’t working hard enough. And there’s a darker inverse of this: the person who constantly fails their goals but fails to realise that the fault is not in themselves, but in their goals.

I’ve been playing with an idea that’s inspired by an exercise in Steve McConnell’s book Software Estimation: Demystifying the Black Art. It involves making guesses about a series of quantities you can’t possibly know, such as the number of trees in Canada or the number of words in the complete works of Shakespeare (examples made up by me, I don’t have the book to hand).

The idea is not to try to a point estimate at the number, but to specify a range that you think is 90% likely to contain the real value. The real challenge comes not in picking a range that contains the real value, but in picking a range narrow enough so that you’re only right 90% of the time.

It would be trivial to say “the number of trees in Canada is between 4 and 7 trillion trillion”, but you’ll be right with probability 100%. It’s a much harder challenge to be right 90% of the time.

Of course, you can’t tell whether a single range estimate has a 90% chance of being right, because the only way to check it is to look up the real value, which only yields a single yes / no. But if you repeat this exercise with 10 or 20 different examples, you’ll start to see a pattern. Most people set their ranges far too narrowly, and are right much less than 90% of the time, but it’s also possible to get the feedback that you are being too pessimistic, and getting too many answers right.

I think something similar can be applied to personal goals: did you achieve 90% of the individual goals you set for yourself? If you achieved 100%, you could probably be aiming a bit higher.

To be clear about what I’m proposing, I think this works best when individual goals are binary (achieved or not achieved), and the feedback mechanism is the number of goals you achieve. In theory it might make sense to accept 90% completion of individual goals, but human nature is such that it just moves the goalposts: if you set a goal to write 10 blog posts but accept 90% completion, then you’ve really just set a different goal of 9 blog posts.

I’ve tried this myself, but not very formally. At one point I actually wrote a goal tracking system that would grade me on how close my task completion rate was to 90%, but I stopped maintaining it. But my experience of using this approach loosely has been pretty positive: it keeps me focused on achieving things but takes away some of the sting of failure if I miss a goal. More importantly, it keeps my attention on the fact that often a “failure” is actually a mistake in goal-setting, not goal-doing. I’d be interested to hear if anyone else has tried anything similar.

Leave a Reply

Your email address will not be published. Required fields are marked *