When Can You Generalize From a Specific Experience? Do We Discount Unusual Events Often Enough?

One of the trickiest things for me to figure out is when I can generalize from a specific experience versus dismissing it as an outlayer.

Take a CEO. He starts his first company. He employs certain management techniques, strategies, and protocols. The company is wildly successful. He moves on and starts his second company. Naturally, he will generalize from his past experience and do what he did before. Maybe it will work out, maybe not. Were the circumstances at the first company representative of circumstances most new companies encounter? Maybe, maybe not.

It is hard, for example, to generalize from successes or failures during the dot-com boom. If you built a company in 1998 and slowly and steadily hired employees and focused on long term viability of the business, you probably would have been deemed a failure, VCs might have ignored you, and so forth. Does this mean you should carry this experience with you as the wrong way to grow a business? Of course not. The dot-com bubble was crazy and unique.

That’s an easy example. It’s harder if we look at the micro level. For instance, one of the very first sales pitches I did (pitching enterprise software to government organizations) was a total disaster. The prospect was unresponsive. The format of the pitch didn’t flow. I could have abandoned it, but fortunately I didn’t. I stuck with it, and it worked out more times than not. As a first-time entrepreneur, however, it was tempting to let any single, new experience have outsize impact on me.

There are a bunch of interesting cognitive biases that are at play here — for example, the recency effect multiplies the impact of recent events versus old events. This is irrational: we shouldn’t let something that happened yesterday seem vastly more important than something that happened last week.

So the dilemma is this: When can we transfer knowledge and experience from one event to another? When are lessons of an experience transferable? I would argue we generally overestimate the transferability of experiences and do not discount unusual events often enough.

4 Responses to When Can You Generalize From a Specific Experience? Do We Discount Unusual Events Often Enough?

  1. Ben Casnocha says:

    Anonymous Commenter: the key is that you need to understand causality. If you don’t really know what caused something, you can’t generalize.

  2. Dave Carlson says:

    That is a very, very good question that really got me thinking. I haven’t figured this out yet, but here are my thoughts.

    In an ideal world, we’d be able to take into account and compare every one of our experiences in a fully organized way. However, this would take far too long and would make us incapable of quick, responsive action. The other alternative is to do something by gut instinct, which enables us to be highly action-oriented but unresponsive to the environment. The way that people are is somewhere in between these two extremes. Most successful people do things by instinct until the results do not align with their expectations. Then, we must reflect and learn from these unexpected events, asking questions such as “what was different this time from the other times that my strategy did work out?”, “Could I have done something different and achieved a better result?”, “Can I look at previous experiences to help me figure out how to better approach this one?”

    The reason people put emphasis on unusual events is that they are jarring, shocking, and unfamiliar. They challenge your way of viewing things, and so you have to put more energy into consolidating these experiences into your existing set of beliefs.

    Honestly, I’d be tempted to say that you can’t come up with a great way to determine whether to generalize from a certain experience, because of the nature of generalization – no two experiences (or states of mind throughout our life) are ever the same, and so we’d always run into certain problems that required constant reflection. I believe that the potential of the human mind to develop is infinite, and the whole essence of growth is that we have to keep asking and mulling over questions like these (as frustrating and time consuming as this can be).

    Then I remembered this sentence from your summary of Emotional Intelligence (which was very well-organized and concise, by the way): “Damasio research suggests that role of feeling is critical to making decisions…emotional wisdom garnered through past experiences. Logic alone can’t decide what job to take or who to marry.”

    The key has to be a good balance of action (in order to obtain new data and information from which to base your conclusions on) and reflection (to analyze and synthesize the information available so you can achieve better results and have more directed actions). It’s a constant question of making new hypotheses about life and testing them out. And I agree with you that we should treat every experience as equally important and an equal part of the “big truth”, and avoid giving any experience more weight (which we tend to do because of these cognitive biases we have).

    I think if we trust our past experiences and act based on what we know, while at the same time accepting what we don’t know, we can’t go wrong. It seems like the only answer is to continue asking questions. (Which you managed to do quite well in this entry :)

  3. krishna says:

    I think one critical factor in this issue of `transferability’ or `non-transferability’ is the adequacy of the “event pool” under consideration.

    Presented with a large body of evidence to relate, we will be tempted to apply the outcome of every event to the given problem to see how close it gets. The larger the event pool gets, higher the probability for many unusual events to group together and present as a `most likely generalization’ .

    With a smaller body of evidence, we are tempted to do a “lean-towards-the-mean” approach – which logically extends to “transferability” norm.

  4. anon says:

    I am perhaps a little late to comment on this post, but as its so interesting id like to add my two cents on a related topic.

    I would be tempted to ask whether culumative experiences help us get closer towards a substantiated form of objective truth. If the answer is yes, then it follows that older people are in better stead to reach the objective truth (unassisted by wisdom – an objective term which works in a subjective field).

    Whilst the ‘lean towards the mean’ approach sounds logical, we should be wary of what JS Mill terms ‘the tyranny of the majority’.

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>