|
VALUE
IMPROVEMENT
LEADERS
|
|
|
|
880 words + 3 activities
| 40 minutes (4 to read, 23 to watch videos, 13 to talk to your coach)
|
|
|
PRINCIPLE
To know where to improve, you must separate the vital few from the trivial many.
TOOLS
Pareto analysis, Excel
APPLICATION
Examine your project for narrowing or prioritizing needs and ask your coach if a Pareto analysis is warranted.
|
|
|
Filtering Trivia As Survival
Your brain
filters out trivial
details full-time in a process neuroscientists call sensory gating. For instance, did you notice the goat woman playing violin above? Of course you did, because it’s non-trivial. Our senses receive millions of details and our brain processes only those which it thinks are important. But you may not have noticed that this article’s font is different from every other email you’ve received this semester, because, who cares?
Giving full cognitive attention to the constant stream of incoming trivia would overwhelm our brains. Processing it would be paralyzing. While we contemplated, say, the helical pattern of a tree’s bark or the variety among its leaves, a puma might pounce on us. Better to filter out the tree stuff and focus on the puma.
We do this unconsciously, constantly. The value leader’s challenge is to be data-driven about what to pay attention to and what to let pass by.
|
|
|
Separate the Vital Few From the Trivial Many
The go-to tool for identifying the vital few is the Pareto analysis, the tag line of which is in the headline. This would likely disgust its namesake, Vilfredo Pareto, because he devised the tool to demonstrate that 20% of Italy’s citizens owned 80% of the land. Those few landed baroni already considered themselves vital, and the unwashed paying them rent were considered less than trivial.
Signore Pareto used his analytical tool to fight the data-driven fight against 19th century inequality.
We’ll use it here to demonstrate the distribution of points in men’s professional tennis.
|
|
|
You can’t tell from the graph, but the top player, Andy Murray (as of early February 2017), has 11,540 points while 1,726 players haven’t yet earned 1% of his points. Defining where the trivial many begins is a judgment call. There’s so much trivia on this graph, it’s not even useful. Excel can only fit 26 of 2,103 names on the axis. A pared down graph would be more useful.
We could define the start of the trivial with a round, arbitrary number (e.g.: 500 points, or top 100 players) but Taro Daniel has 502 points and Ernesto Escobedo has 496; not a meaningful difference. Likewise there’s only a 3-point difference between player 100 and 101.
Turning back to the rows and columns, I found a significant 15% gap in points between the 23rd ranked player (John Isner, 1,715 points) and the 24th (Gilles Simon, 1,495 points). That’s the last double-digit difference until we get to Tommy Haas, ranked 1022nd. I chose this non-trivial difference to define the beginning of the trivial players. Sorry Gilles.
Here's the revised, now useful graph.
|
|
|
We can make some observations:
- Two players are dominant, in a group unto themselves.
- Players 3-6 appear to form a group and they have collectively earned fewer points than the sum of the top two players.
- 25% of all the points earned went to the top 13 players, leaving the bottom 2,090 players to duke it out for the remaining 75% of points.
- The tail goes long and flat after John Isner.
You’d be right to suggest all those observations are also trivial, because we're not in the pro tennis business.
|
|
|
The Vital Few Driving Decisions
Let’s say you’re leading an effort to reduce process defects. 12 varieties come to light from a cause-effect brainstorming session. Should you take on all 12? Jack Sock (I’m using tennis names) thinks you should prioritize. Grigor Dimitrov agrees and is quite vocal that defect A is the biggest problem. Andy Murray, in an erudite tone, says defect D should be addressed first. (The team should agree; he has the most points. Or is that
HiPPO
?)
Then Luigi Sorrentino (rank: 1,865) reminds the team of his countryman’s analytical tool for guiding a team’s focus. After a query and some data cleansing, Luigi produces this graph:
|
|
|
Some non-trivial observations the team can actually use:
- Varieties B and C account for 50% of all defects and occur in similar frequencies.
- J appears to be a group by itself and brings the cumulative to 66%.
- The least common varieties (n=9) collectively make up 34% of all defects.
I hope, at this point, Luigi reminds his team that “data-drives decisions, but people make decisions.” The team would be wise to consider:
- Feasibility of prevention: It’s easier to prevent problems we cause ourselves rather than those caused by another department or by patients.
- Severity: Defect C may cause our employees minor inconvenience while defect D causes injury.
- Likelihood of discovery: Defect G might lay undiscovered for days causing havoc downstream while defect D is known as soon as it happens.
|
|
|
LINKS
Quickly locate all course videos, slides, and previous emails
here
.
|
|
|
|
|
|
|
