On Measurement 1: Measures

Strategic Preparation Series

“You cannot improve what you cannot measure.” -Peter Drucker (or maybe Edwards Deming)

“Not everything that can be counted counts, and not everything that counts can be counted.”  -Albert Einstein (or probably William Cameron)

While researching these contrasting statements to open this piece, it turns out neither of them may be attributable to anyone. Sometimes I wonder if every maxim by which I live my life might just be a lie.  Oh well, onward! 

Let’s kick off the Strategic Preparation series – my amalgamation of multiple planning and goal setting frameworks into a method for aligning organizations to solve complex problems together, repeatedly over time, which I have found useful for every business I’ve advised. 

In this series, we are going to touch on works I’ve referenced before such as Playing to Win, OKRs, Scrum Agile, and Coordination Headwind. Before we dive in, let’s get oriented around an approach to measurement in complex organizations that is the basis for goals, OKRs, performance evaluation, continuous improvement, KPIs, health measures, etc.

Measurement is a topic that seems to piss off pretty much everyone. The Drucker camp prefers lots of measurements, ideally quantitative, and everything that cannot be measured is just an annoying, messy part of life that can be safely ignored.  The wacky logical conclusion of the approach is the McNamara Fallacy.  We all know people like this, and while I appreciate their attempt to gain mastery over the world, they drive me nuts.  

On the other hand we have the Cameronites, perhaps erroneously waving the Einstein banner, who understand that life is bigger than it seems. Perception is not reality, rather, as Philip K. Dick reminded us, ““Reality is that which, when you stop believing in it, doesn't go away.”  However, the existence of unmeasurable important things does not obviate the utility of measurement. Just because you cannot (should not?) flush your garbage down the toilet, does that suggest that you get rid of your toilet.  

Let’s agree that measurement is an incredibly useful, yet limited tool.  Like all tools, measurement must be applied with discipline and skill by leaders to gain purchase over the unknown.  In subsequent posts we will get into measurements by company stage, how to set goals across any periods of time, how to measure your current reality to ensure reliability and many other uses.  For today (and next week, I’ve been advised to split this up), let’s approach measurements as three distinct, interrelated parts, each with a form and matching purpose: 

1. Measures: the description of what you want to know more about (what and why)

2. Metrics: the specific definition of what you plan to measure (how and when)

3. Targets: the actual “goal” state for the metric, that which you are driving toward 

Measures

Measures are the answer to the question of “what do you want to see, and why?”  More revenue, more patients, better health outcomes, higher quality, shorter wait times, fewer defects, shorter sales times, better engagement?  

Measures are ideas, notions, concepts. They need not be specific enough to actually be measured (the metric’s job) because measures are a communication tool to align your organization toward the work ahead. In OKR language, the “Objective” is the same idea here. Inherent in measure statements are desired outcomes, such that the concepts of goal and measure are somewhat indistinguishable.  There isn’t much reason to measure something unless you want it to change or know when it is breaking.  So the act of measurement is the exertion of agency over reality. 

Measures are easy to create because they are not specific, which is why they proliferate wildly, paralyzing organizations with chaos masquerading as order. As a senior leader, your job is to choose very few measures on which to focus the organization.  You must also work out the dependencies between desired measures to avoid the awkward tradeoffs that your middle managers will have to make.  

For example, let’s say you want more revenue, a classic goal.  Your company could focus on more revenue per customer OR more customers, or both!  If you were to frame the measure as only more customers as a shorthand for revenue, you’ve now biased your teams toward one set of solutions that may or may not work for your business. Even worse if you have more customers AND more revenue as parallel measures, even though there is a dependency, you will confuse your teams and open up all sorts of gaming.  Finally, if what you actually want is more customers, do NOT oversimplify to more revenue or you will drown in rejecting distracting solutions that bring revenue but not customers.  Only senior leaders can ensure this clarity through specificity of measure and as you can see, it gets incredibly complicated incredibly quickly.  I’ve repeatedly seen senior leaders try to push the resolution of this ambiguity down the organization and it always leads to frustration and failure.  Remember, your people will inevitably execute against whatever you tell them, whether or not it makes sense.

Fun, right?  Next time we will pick up with Metrics, the specific definition that makes measurement possible.