Thank You Vilfredo
Vilfredo Pareto publicized an observation in 1906–that 80% of Italy’s wealth was in the hands of 20% of its citizens. Applied at a broader level his observation indicates that 80% of the affects come from 20% of the causes.
I want to be clear; the Pareto principle is a rough guide about distribution–not a hard and fast rule. Since 1906, when the man with the most alliterative name in the history of Economics made his observation, Vilfredo Pareto’s principle has been applied to everything from employee performance to software development.
In very succinct terms related to software, the Pareto principle observes that
- 20% of defects cause 80% of our problems
- 20% of the features get 80% of the usage
But our own observations show that not to be the case. The Pareto principle is a verifiable, repeatable observation that distribution is decidedly lopsided. Sure the numbers may get a little muddled, but–as in everything with Agile–it’s about having the conversation and making sound decisions based on what we can observe.
So, how can we apply it in software development ? Here are a couple of suggestions:
- If 20% of bugs contribute 80% of crashes, then we should focus on fixing these bugs first.
- If 20% of customers contribute 80% of revenue, then we should focus on satisfying these customers.
In other words–we need to be more deliberate about working smart instead of working harder. Again, this goes back to the conversation Product Owners should be having with their customers and with themselves. Catherine Louis has a really great article on the Scrum Alliance site about applying the Pareto principle to the backlog. My blog here is more about getting that conversation started. That’s where it gets a little tricky.
Q: What about the 20% of the users who log 80% of the complaints? Should we focus on them? ( I mean, making the complainers happy will lower my stress, right?)
A: Not really. Responding to issues costs money, and responding to issues raised by non producers should not be a priority, and. . . and OUCH! That kinda stings, doesn’t it?
Pure and simple. If those complaints are from customers who are not producing revenue, then the work to fix those issues has to be be prioritized accordingly–i.e. lower. That’s right. We don’t get to make everyone happy. It may seem cynical, but if your company’s goal is to make money then you should focus on helping where the money is. If you are in business, and you think your job is to make everyone happy, you are destined to burn out very quickly. On the other hand, you will always bear positive results if you focus on the people who are currently helping your company make money and on those who have the potential and demonstrated willingness to cooperate.
Harsh? Yeah. a little. Pragmatic? Definitely.
Try reasoning through it from the opposite direction:
Let’s say, I have 100 users of my system, one of whom is “George.” 80 other sales people are able to use the software successfully. George is consistently in the bottom 20 of my revenue earners. George takes up 80% of my time, because he doesn’t like the software. Conclusion: I should spend my time changing the software to make George happy and ignore the needs of the other 80.
Sorry, George. That is an irrational conclusion for me to come to.
The point of talking about the Pareto principle, (or at least keeping it in mind when trying to decide what changes would be the most effective), is to acknowledge that distribution is always lopsided. Your real task is to identify how it is lopsided and make decisions accordingly.
Another great little site, Better Explained, literally illustrates the principle, (and does so very well). That same site also sums it up the other side of the point nicely
“don’t think the Pareto Principle means only do 80% of the work needed. It may be true that 80% of a bridge is built in the first 20% of the time, but you still need the rest of the bridge in order for it to work. It may be true that 80% of the Mona Lisa was painted in the first 20% of the time, but it wouldn’t be the masterpiece it is without all the details. The Pareto Principle is an observation, not a law of nature.
When you are seeking top quality, you need all 100%. When you are trying to optimize your bang for the buck, focusing on the critical 20% is a time-saver. See what activities generate the most results and give them your appropriate attention.”