Key Metrics

Two Cal Newport articles, one from 2009 and one from last month, both tackle the topic of narrowing focus to a pinpoint and only considering feedback specific to those specific goals.

An excerpt from the first post (which I highly recommend reading):

We soon hatched a plan. He would forget the random advice he’d been receiving from various friends and hangers-on — the suggestions to hand out demo CDs in front of radio stations or network to meet the right executives. Instead, he would turn his focus solely onto the Pyramid Club. He would return to the open mike again and again until he was able to win over that crowd. After that, he would progress to the showcase and play to win.

Our logic was simple: if he couldn’t become good enough to win over the Pyramid Club crowd, he couldn’t become good enough to attract industry attention. So why waste energy doing anything else?

And the second, more recent post (also a great read):

My above experiment sweeps these compelling sounding ideas off the proverbial table, and replaces them with an approach backed by data. What matters, it tells me, is something we can call: quality cited papers. In more detail, how many papers per year are you publishing that: (a) are in quality venues; and (b) attracting citations? 

This metric can tell me if I’m improving or not from year to year. Similarly, it provides clear feedback on which of my research directions should be dropped and which emphasized. When deciding whether to join a project, for example, I should start by estimating the expected impact on my quality cited papers value for the year. When deciding whether to apply for a particular grant, the same question should guide the decision.

I take the combined message as follows: be extremely deliberate when it comes to goal-setting, and be just as deliberate about choosing key metrics for tracking progress. We want to have as few key metrics as possible, and we want each one of them to be practically synonymous with goal achievement.

Let's take strength training as an example to illustrate the point. Say I want to become stronger, so I buy a gym membership and starting lifting weights three times a week. Pretty soon I devise a training plan and start tracking my performance on a variety of different exercises: pulldowns, biceps curls, leg extensions, dips, crunches, etc. Each workout I attempt these exercises in whichever order the machines are open, sometimes but not always making it through the entire list. Generally my performance improves, more or less, but the gains are uneven and hard to decipher. Sometimes I do so well on the chin-ups that I can barely perform a curl, and other days I save my favorite exercise for last but find that I've completely run out of energy to do a single rep. It ends up being pretty discouraging because I seem to be moving backwards at something every single workout.

Contrast that with the "key metric" approach: the only metric that I'm going to monitor closely will be the squat. Squatting is the most intense and important full-body strength exercise there is; being able to squat a lot of weight is practically synonymous with strength. You can't squat 400 pounds with a weak core, inflexible legs, or an underdeveloped chest, arms, or posterior. Everything that I consider doing within a strength-training context will be judged against the question of "will this help me increase the amount of weight I can squat, or not?" So, yes, I'll continue to work on increasing my bench press and chin-up numbers, but the squat is the only exercise that 1) I will consistently perform during every single workout, at the very beginning of the lift session, and 2) I will mercilessly demand a personal record from myself each and every time. If the numbers keep going up, then my strength is undeniably improving; all other metrics are just noise.

Attempting to track (and thus allow yourself to potentially panic - or gloat - over) every single quasi-relevant metric is like congratulating yourself for ordering a Diet Coke to go with a 1,200-calorie large buttered popcorn. The key metric is "calories-in-calories-out", not "fructose calories avoided." Remember that in a dynamic system anything that optimizes the sub-parts tends to sub-optimize the whole.

Here's a couple other examples of Key Metrics:

• Cardiovascular fitness: Is my five-mile run time decreasing month over month? NOT: how tired am I after walking up stairs, how fast can I run 100 meters, am I groggy in the morning...
• Work output: Am I increasing the number of high-quality projects I complete each month? NOT: how many hours am I spending in the office, how many random compliments or kudos I receive during the day, how large my most recent raise was...
• Pretending for a second that I have my dream job as an NFL General Manager: Are we increasing the ratio of production to salary expenditure over time? Notice how the very best NFL franchises tend to eschew pricey free agent acquisitions in favor of value and upside. The Baltimore Ravens' GM is clearly not measuring himself against the "how many aging veterans can I re-sign to placate our fanbase?" metric.

Contrast that approach with the Philadelphia Eagles over the past few seasons...