# Switcher’s curse and legacy decisions

Switcher’s curse is about conservatism in a world of constant change. There are many fallacies and biases which cause us to stick to legacy decisions. Switcher’s curse is a a fallacy that causes us to switch when we should stay. Keep reading about the challenges of legacy and how to avoid catching the curse. Keep reading for a background to the problem or jump directly to my explanation of Switcher’s curse.

## Progress at a breakneck speed

Research and development is progressing at a breakneck speed. Companies, products and ideas are becoming shortlived. We live in an age of change. Sticking to the old is foolish. This is the voice of our time. More often than not, switching to something new is better. A newer car consumes less (zero?) fuel and costs less to maintain. Sticking to old operating system versions is fool hardy.

We have all been warned about the “sunk cost fallacy“. Freakonomics teach us about the upside of quitting. We should avoid emotional attachments to old decisions. We should look at future net benefits and switching costs. Based on that we should make rational decisions between staying on course and switching. But not only do we need the best information, we also need the best decision models.

## Switcher’s curse

Look at these dolls. They are cursed somehow. Perhaps with the Switcher’s curse.

[bibshow]”Switcher’s curse” is “a trap in which a decision maker systematically switches too often”. [bibcite key=”citeulike:14027879″] Consider a situation where a decision maker at some point in time has to decide: should we stay with A or switch to B. A is the incumbent policy or system or whatever. B is the new opportunity. Let $g_A$ and $g_B$ be the goodness, e.g. NPV, of A and B respectively. Let $s_{AB}$ be the switching cost from A to B. Consider a decision criteria $d=g_A-(g_B-s_{AB})$, i.e. are the net benefits of staying higher than of switching?

1. If $d>0$ we should keep A.
2. If $d<0$ we should switch to B.
3. If $d=0$ we can just flip a coin.

This model is to naive, writes Edlin. It doesn’t take into account that g and s are just estimates. A different sampling would have given a different outcome. We can see that $d\gg 0$ to be on the safe side before we switch. We can quantify this as $d>m$ where m is a safety margin.

Because A is the incumbent system, it has been subjected to this kind of choice before. We might not know or remember why the decision was always made to keep A. We do know that the longer we have kept A, the more often we have selected to not replace it. That means that m grows over time. In other words, switching has to be much much better than staying before we should take the risk to switch.

## Existing software embeds decisions

Existing software is much more than a series of decisions to keep it. It embodies a huge number of decisions on how the business of the company should work. The software is full of decisions about business objects and what should be done with them. These decisions, embodied in the software, forms the operating system of the company. The decision to switch is bigger than replacing some immaterial asset with another. It is a decision about replacing a proven way of working with a new way of working.

Switcher’s curse is real. Conservatism is often the best approach. But it needs to be a prudent conservatism. Making changes smaller and more easily reversible decreases the need for caution. We should consider a prudent application of fail fast mentality in our decision making process. (But I prefer to call it learn fast.)

We don’t always know why things are the way they are. We should always assume that there is a good reason that they have not changed in a long time. The grass is seldom greener on the other side. Where ever you go, there you are.

## Part of an ongoing conversation

This blog post is part 18 of an ongoing conversation. In part 17, Gene Hughson discussed how what is “old can become new again“. I hope you agree with me after reading this post that “old isn’t always old”. Gene maintains a list of all the posts in the innovation series.

[/bibshow]

Image sources