Measurement and Management

Over the years this blog has evolved, as my job has changed.  From software delivery manager, consulting architect, to enterprise business architect.  It has often been about management and leadership because those topics have been at the front of my immediate list of problems in delivering software.  Today I read this thoughtful post… by Henry Mintzberg that arrived in the form of a retweet from a respected blogger and agilista.

In the post Henry pretty much destroys the old aphorism “If you can’t measure it, you can’t manage it.”  Except even after reading his rather thorough diatribe, I disagree, and I don’t think he actually offered any suggestions for how to manage things that “can’t be measured”.  In fact, what he seems to be railing against is something that he says way towards the bottom – which is measurement as a substitute for management.

So lets unpack what he is saying:



v. man·aged, man·ag·ing, man·ag·es


a. To have charge of; direct or administer: manage a company; manage a portfolio of assets. See Synonyms at conduct.
b. To exert control over; regulate or limit toward a desired end: manage the news to minimize political repercussions; managed smokestack emissions.
c. To direct or supervise (employees or other staff): She manages 20 people in the department.
d. To act as the manager of (a performer, for example).
2. To succeed in accomplishing, achieving, or producing, especially with difficulty: managed to get a promotion; managed a polite goodbye.
3. To succeed in coping or dealing with: a drug that improves patients’ ability to manage their disease.

1. To direct or conduct business affairs.
2. To continue to get along; carry on; cope: learning how to manage on my own.

My first problem…

with Henry’s attack is that he didn’t clearly define which of the definitions of manage he is attacking.  But behind that, lets say I select definition b)  To exert control over; regulate or limit toward a desired endI think that when most of us think about managing work – this is what we mean.  I know this is semantically picky, but you can direct people but manage work.  Managing people implies directing or supervising someone’s activities.  As a manager, you are not managing their personality, their emotions, their motivation, their values or their attitude.

My second problem…

is that Henry uses a bunch of examples of things that he claims to be hard or impossible to measure, so I want to ask him for suggestions for how to “manage” them in the absense of being able to measure them…
Culture – Not only is culture hard to measure, it is notoriously hard to manage.  On of my favorite aphorisms is “Culture eats strategy for lunch.”  The way I think of culture, it is the combined value system that an organization invests in.  You can’t manage it, you can only influence it.  Certainly some individuals have more influence than others, but culture is a social thing, and no one person controls or directs it.  Of course, the person who controls who is in the organization has the most influence by hiring people who already have invested in the desired values and by separating people who demonstrate contempt for those values.
Markets for novel products – Markets for novel products are notoriously difficult to measure, or even to comprehend.  But can anyone actually manage a market?
The performance of management (or managers) – This is less difficult than it may appear.  Management (as the group of people at the top of an organization) have financial measures of effectiveness.  The stock price is an interesting measure, but there are so many variables that influence it some are clearly beyond the control of the board room.  But there are other financial performance measure in play.  There are also the measures of execution, the ability to steer and the ability to accomplish organizational transformation, the ability to achieve stated goals.  Management (or more properly leadership) clearly have measures of performance and measures of effectiveness.  Even managers (lower level managers) have accountability for the output of their staff, so there are clearly measures of effectiveness or performance at all levels of management.
Learning – Can one measure the ability of a person to learn?  Of course.  Can on measure the amount of knowledge out of a finite knowledge base that a person has assimilated so that they can apply to solve specific problems, or accomplish specified tasks?  Yes, absolutely.  However, the way tests are administered in the US, the schools actively participate in shenanigans in an attempt to game the metrics by “teaching the tests” and other such folly that make the whole “education process” much more a a waste of time than would otherwise be the case.  Mostly this is caused by underfunding the system, and using composite test scores to punish the schools that likely need the funding the most.  This is a clear case where the metrics have been weaponized.  From my view, the testing that is done in school should be diagnostic, rather than measures of performance.  That is, used to determine which students need more support, or lack capacity in certain areas.
In this regard, I pretty much completely disagree with Henry.  I think that the things that can’t be measured it is hard to know whether you are even making a difference, so while you can exert influence, without measurement can you actually tell whether you are being impactful, and as soon as you point to something that demonstrates impact – you have identified your measure, right?  How can it not be so.

Where we agree…

What I agree with is a sort of obsession with measuring.  There is a sense that if measurement is good, than more measurement is better.  But measurement is not without cost, and often it brings with it a drag on our ability to accomplish the work that is being performed.
I also think that we give too much meaning to measures, and so there is a tendency to “game the metrics” to appear to be outperforming.  Or on the other side, to “weaponize the metrics” and use them to beat employees or worse, prevent them from being fairly compensated.
I think that there is a difference between measures of performance or measures of effectiveness and diagnostics.  Where as the measures of performance or effectiveness indicate quota or a bar that must be surpassed, diagnostics tell us if we are having a material impact, if we are “moving the needle” in the right direction.
I would say that based on all of the definitions of management above, what managers need are diagnostics more than measures.  Diagnostics are used to help managers know that they are having the right influence, or an expected impact.
One last point – there are differences between what have been called “lead measures” and “lag measures” – Lead measures are indicators that an organization is doing the things that it believes will ultimately lead to positive results.  Lag measures are the measurement of the results themselves.  In most cases, by the time the results are “in” – it is too late for managers to “do anything” about it.  Steering by lag measure is like driving by looking in your rear view mirror.  Hard to see where you are going, when you are always looking where you’ve been.  Lead measures are much closer to diagnostics they are the mile markers and the compass needles that tell us we are making incremental progress towards our destination even when it is dark outside and we can’t see the surrounding landmarks.
So Mr. Mintzberg, thanks for the stimulating TWOG or whatever the heck you call it.  It really got me thinking.

No Comments

Leave a Reply