Using Prospect Theory to Guide Product Management Decisions

In working through a number of recent product decisions, it quickly occurred to me that the fields of prospect theory and product management are linked together. In that customers (and internal employees) perceive removing vestigal, unused features as being far worse than never developing it at all. Here’s a breakdown of why that is, using the field of prospect theory as a guide.

Prospect Theory in a Nutshell

Prospect Theory won researchers Kahneman & Tversky a Nobel Prize, and comes up in everyday terms as “loss aversion“. The basic idea is that people value losses much more harshly than gains. So if you lost a $100 bill you would will lose more satisfaction than you would have gained from finding a $100 bill.

The way that looks graphically is shown in the picture – the asymmetry in the curve illustrates that with the same value results in a difference whether its a loss or gain.

Fore more info here’s the paper: Kahneman, Daniel, and Amos Tversky (1979) “Prospect Theory: An Analysis of Decision under Risk”, Econometrica, XLVII (1979), 263-291.

Implications for Product Management

Clearly, it seems that “taking something away” from your customers is perceived as much worse than the reception you would get by giving it to them initially. So, as a data driven product manager, what can you do when you realize that there are features of functions in your product that are no longer use – or the cost to maintain are not worth the effort.

Here are some examples:

• First of all, introduce features slowly and carefully to avoid the situation altogether
• Instead of removing the feature entirely, rebrand and bundle it into an new feature
• Make it less accessible, or make it available on-demand, or as a service to phase it out
• Give lots of notice and sufficient time to readjust customer use patterns — also so that introducing new features minimizes the impact of the loss (i.e. people will have forgotten about it by then)
• My favorite, but most controversial, is to just cut it without asking and see who screams and how loud…chances are if enough people complain, they’ve just helped you get closer to understanding the optimal value, and you’ll put it back in (with hopefully an upgrade)

The point here is not to prescribe a particular solution, but to help recognize the underlying behavioral pattern so you can address it proactively or respond promptly and painlessly.

Like it? Share it!
Shortlink: http://bit.ly/prospecttheory