Learning from Failure, learning from Success

---

Trial and error, or trial and success? Which is the better learning mechanism? 

Crossroads by Chris Potter on Flickr and stockmonkeys.com

A thought provoking piece here, makes the argument that people learn much better from their mistakes, as a result of the emotional charge and the emotional scars that failure brings.

On the other hand, a well-argued article in  Business Insider suggests that success is a better teacher, because a fail mode only tells you one option not to try; it does not tell you how to succeed.

Which is correct, and what is the implication for Knowledge Management?

Personally I tend slightly more towards the "learn from success" argument.

I believe that on a corporate level, or as a society, we collectively make the biggest learning steps when we finally succeed. Think of Edison and his light filament; when did he do the most learning? When he tried each of the 99 options that didn’t work, or when he found the one that did?

Now you can argue that he learned from both, but now let’s look at transferring that knowledge. If you were a light bulb maker, which of these two statements would be of most value to you?

1. You can’t make a light bulb filament out of cat hair 
2. You can make a light bulb filament out of tungsten 

The implication for Knowledge Management

The implication for Knowledge Management is this:

You need to learn from both success and failure, but you need to learn much more carefully from success, because success is what you want others to replicate.

It is learning from the successes that are most valuable for others, so that is where the most KM effort needs to be applied.

As an example, let's look at the typical systems set up to learn from safety incidents. Most of these systems have detailed root cause analysis when there is a near miss or an incident, and the lessons from these are sent around the organisation so others can learn from this safety failure.

However it is far more important to learn from the factory or plant that never has an accident, and never has a near miss. That is the factory everyone needs to emulate, which means they need to carefully understand WHY they are so safe, and then learn from this.

We know that it is human nature to learn best from mistakes, but we don’t want to be at the mercy of human nature. We don’t want people to have to screw up in order to learn. We don’t want failures and screw ups if we can possibly avoid it, because mistakes and screwups can cost money, they can cost lives (in certain cases), and they can cost careers if they are big enough.

Learning in a KM Framework

The ideal situation, in any mature KM or lesson-learning framework, is that you learn as a matter of course, whether you deliver failure, success or a mixture of both (and it usually is a mixture).

Learning, as in lesson identification meetings such as After Action reviews and Retrospects, should be a routine exercise, regardless of success or failure. Address both, learn from both, and give particular attention to understanding the causes of success.

Fail fast and fail often is a good mantra, so long as it leads to success, and it is the success that is the greatest learning opportunity.