Why Can’t We Figure Out How to Select Leaders?

By Michelle Malay Carter on February 16, 2009 

Why Can\'t We Select LeadersJim Heskett at Harvard Business’ Working Knowledge has another question up for comments:? Why Can’t We Figure Out How to Select Leaders?

My Answer is Simple
It is because we don’t understand work or the variations in humans’ ability to perform work.? When we try to match leaders to jobs, we are like 18th century doctors trying this, that, and everything to cure people with no clue about viruses, bacteria, or cancer.? Some things work some times.? We grant causality to interventions that were not causative, etc.

Below is my response to Jim Heskett’s question.? Much of it has already been articulated on this blog, but since the question is still being asked by our top thought leaders in the field,?I thought I’d try yet another time.

Why Can’t We Figure Out How to Select Leaders?
The reason why we can’t figure out how to select leaders is that we don’t understand work. The reason why I can diagnose this problem with confidence and ease is that I have a model – a model for work and a corresponding model for human cognitive capability to perform work.

Management Science Must Take a Page from Physical Science.
An analogy: An understanding of work levels and its relationship to levels of human problem solving capability is to management science what the understanding of energy and its relationship to temperature is to physical science.

Simple Tools and Engineering Templates Allow for the Easy and Practical Application of Scientific Theory
When the thermometer became a tool for the easy, practical application of scientific laws toward predicting the “behavior” of matter, our ability to engineer our physical world to our benefit improved dramatically. For example, not only could we now observe that H2O existed in three different states (ice, water, steam), the thermometer allowed us to reliably forecast when these states would change.

If we understood work, we could use this knowledge to engineer structurally sound organizations and to hire leaders who match the work required by the role.

Back to (Little Known) Management Science

  1. Not all work is the same. Work exists in discreet levels of complexity, and any given work role can be categorized by level. Time span of discretion allows us to measure this.
  2. Raw talent capability to solve problems in humans exists (independent of experience, education, or knowledge) and it can be categorized by level. Managers, given a common language and framework for interpretation, can reliably judge their employees’ capability levels.
  3. Human problem solving capability levels can be aligned one-for-one with work levels.

Matching People (And Leaders) to Roles
If we understand the work level called for in the position of CEO (which will vary from organization to organization), we can conduct a selection process that targets and screens for leaders with the requisite level of cognitive capability.? Just like we can observe H2O in its various states, we can also observe and judge cognitive capability.

Unfortunately, if the CEO role is at level five and a leader currently capable at level four is hired, s/he will shrink the organization down to something s/he can contain as s/he will not be able to perform the level five work. ?If a CEO with capability above the role is hired, brace yourself for growth.?? More on this phenomenon here.

One Piece of a Three Point Model
Cognitive capability is not the only criteria to consider, but it should be the first. I use a three point model to match people to roles. One point is cognitive capability. Once you have established a candidate matches a role in terms of cognitive capability, one must also consider knowledge, skills, experience and values, temperament and inhibitors.

Managerial Leadership Tools and Organizational Engineering Templates Based on Scientific Theory
Tools do exist for the easy practical application of work levels theory toward predicting the behavior of leaders. Hence, our ability to engineer our organization’s systems to tap all employees’ full potential is one benefit of using work levels theory to design work enabling systems. Other applications of the theory can benefit organizational effectiveness in a variety of ways.

Meta Model Requisite Organization
Further, work levels is just one piece of the meta-model, Requisite Organization, which offers an integrated approach for organization design, managerial leadership, and talent management. It’s not rocket science; it’s people science. I suspect the job title, organizational engineer, will be mainstream one day. As I like to say: I’m OK. You’re OK. Let’s fix the system.

Need Proof, Want Research?
If you would like to read more on the management science part, you can access the 210-page Requisite Organization annotated research bibliography or 1010 pages of dissertations on the theory sponsored by the Global Organization Design Society.

What Will It Take?
Interestingly, Van Leeuwenhoek’s discovery of bacteria in 1683 was not immediately accepted by scientists. His letter to the Royal Society announcing the discovery of bacteria caused such doubt at the Royal Society that he had to enlist an English vicar, as well as jurists and doctors, to confirm that his report was based on true observations.

When will you believe the discoveries that have already been made in the field of management science and use them to your benefit?? I’m OK.? You’re OK.? Let’s fix the system.

Filed Under Executive Leadership, Requisite Organization, Talent Management, Work Levels


4 Responses to “Why Can’t We Figure Out How to Select Leaders?”

  1. Wally Bock on February 18th, 2009 5:37 pm

    Looking only at first line supervisors, we’ve got a different situation than you describe. We know pretty well what kind of person will make a good and effective supervisor. He or she will be hard working, of course, and smart enough. Beyond that, if we promote people to the supervisory job who enjoy helping others succeed, are willing to talk to other people about performance and behavior and who can make a decision, the odds are good that we’ll get an effective supervisor. The next task is giving him or her the tools and support to succeed.

    Despite the fact that we know this, we promote people whose primary qualification is that they’re good at the work to be supervised. Then we hope they’ll do well and we’re surprised when many of them don’t.

  2. Michelle Malay Carter on February 19th, 2009 7:51 am

    Hi Wally,

    Yes, a candidate has got to have the whole package. Values and temperament are extremely important because if it’s not natural, it’s not sustainable. So, if I don’t care to supervise others, I am going to dread coming to work each day and avoid the supervision part of my work. Not a good situation.

    Thanks for the comment.


  3. Erik LaBianca on February 23rd, 2009 10:49 pm

    Hi Michelle,

    You mention scientific tools to measure time span of discretion. Jacques on book on the subject is now out of print, and his techniques seem to be pretty obtuse (interviewing and analyzing the conversation.

    I’d love to hear what sort of techniques you use.



  4. Michelle Malay Carter on February 24th, 2009 4:45 pm

    Hi Erik,

    Thanks for stopping by and for the question. Time span of discretion of a role is in the mind of the manager with the approval of the manager once removed. So, yes, you have to interview the manager about the longest task the manager is holding the role accountable for. It takes some training and experience to tease this out because most managers don’t think in these terms.

    Maybe I’ll do a post on the subject.