Forecasting by Analogy

I recently came across a piece by Bala Iyer on titled, “To Predict the Trajectory of the Internet of Things, Look to the Software Industry,” and I wanted to call it out for readers as a good example of what we call “forecasting by analogy.” It is something most people do when thinking about the future, and all of us probably do it more often than we realize. It can a very useful way of forecasting changes, yet we want to always be careful in relying on analogies.

Applying Patterns


Figure 1: VFS icon for theories of change

In essence, forecasting by analogy means identifying a pattern of change that happened in one subject or area in history and applying that same pattern to a new phenomenon or issue. In past years observers have used analogies from things like the development of radio and television to forecast (some aspects of) the future of the internet, or how the outsourcing of jobs in one industry might presage the future pattern of outsourcing in another industry. In the above article, Iyer is applying a specific set and sequence of changes he observed in the development of the computer software industry to anticipate the changes that will take place in the evolution and maturation of the Internet of Things.

Serious analysts like Iyer will do a lot of research to understand the intricacies of what happened in the historical example they are applying, but most of us tend to be very quick in applying historical patterns. Forecasting by analogy is about patterns and humans are natural pattern recognizers and incorrigible pattern completers. Thus, we have an inborn reflex for applying experiences (read: patterns we’ve seen) to future uncertainties. Not a bad way to go, given a pre-historical hunter-gatherer lifestyle living among saber-tooth tigers and giant sloths where snapping twigs in the dark probably meant some creature up to no good for you, but forecasting by analogy can lead us astray in the complex, modern world, particularly when we do it without much thought.

Being Careful When Applying Patterns

First of all, keep in mind that typically when we are using an analogy to forecast future uncertainties, we are using a pattern from a different phenomenon. That is, we are using the analogy becomes things seem similar to us (radio seems kinda like TV, which seem(ed) kinda like the early internet; outsourcing in one industry can look on the surface just like outsourcing in another industry). We need to keep in mind, however, that when we do this we are taking a definite leap of faith that how things work in one area relate to how they work in another.

Second, don’t forget that relying on the known history of one thing (there’s some form of data there) can give us the false sense of comfort that we are basing our forecast of the other thing on “data.” If the thing you want to forecast is extremely new to the world and doesn’t have a lot of its own experience (data), then you could end up feeling good about your forecast for the wrong reasons.

So, what can you do to help with the forecasting by analogy that you might want to conduct? If you have more than just a passing interest in how a new emerging issue or technology might turn out, then consider the following guidelines for forecasting by analogy in your daily work.

  • Keep in mind that the analogy you apply is really about how you are framing the issue. It ends up defining what you believe drives change and what interactions you think are most important. It is the first step in filtering the signals around you. This is where blind spots begin to form.
  • In many cases, your forecast by analogy may only represent your first attempt at anticipating the future of your issue. We use this method because we think (hope) that other events in history provide useful insight into the future of another issue. Especially if you’re not going to be conducting hardcore analysis, you should think of your forecast by analogy as your “first pass” at anticipating the future.
  • Be extra vigilant to question your own assumptions about how things are going to change. Because your forecast by analogy is based on what might be historically contingent circumstances (read: may have only happened once because of very specific events that are unlikely to repeat in the exact same way), you will need to continually watch for signals that indicate your issue is taking a different path through history. In essence, you want to keep watch to see if the analogy does not hold true.

Patterns Can Be Fun!

Most of us will continue to do a lot of forecasting by analogy in our lives, so hopefully the suggestions listed above will help to make your forecasting more rigorous and robust. This kind of forecasting, particularly when done in a quick, everyday sort of fashion (as opposed to serious research projects that dig deep into history) can be very useful for helping people develop some initial handle on an emerging uncertainty, and it certainly can give people the confidence to take further action. We just want to be sure that you are conscious of the pitfalls and potential downsides to the method.


To Predict the Trajectory of the Internet of Things, Look to the Software Industry.”


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: