Myths of Measurement: Is Performance a Real Thing?

When I was a child, a teacher was considered high performing if he had a quiet and orderly classroom. This is no longer true. As pedagogical theory as evolved, student engagement in learning has become more important than order and quiet. Now, if children are noisy but engaged, a teacher is performing well. Which model of teacher performance is correct?  

In the last 40 years we’ve seen two different models of teacher performance.  Our understanding of employee performance evolves. Actually, our understanding of talent, which is largely conceptual, is constantly evolving, and varies between people. In terms of teaching, discipline and order were important; now engagement is paramount. This is not an obscure pedagogical point. It is the key to successfully using talent measures.

Mental Models of Talent 

The most important features of talent are invisible—features such as performance, potential, personality, and intelligence. As a result, we have mental models of talent. Our ideas of talent and performance change over time, as we saw in the example above. In a practical sense, everyone agrees that a desk is a desk, or a rock is a rock, but performance is not always performance. 

Consider sales representative performance in the life insurance industry. The measure of sales productivity is commission, and commission is a percentage of the premium paid each year for a policy.  A veteran salesperson in this industry may not be selling new policies, but may be paid handsomely for policies she sold years ago, since the customer is still paying the premium.  This is a unique model of performance, since it includes results of behaviors from years past.  It is quite different from a typical model of sales performance, which is the amount of product sold in a month or quarter. 

Clearly, performance is not simply performance. Apparently obvious measures of performance, such as sales, involve assumptions. Even the idea of sales performance is a model. In a business context, the mental model of talent or performance is built by management’s expectations.

In a general sense, models of talent are networks of theories and assumptions. It could be a theory about how people tend to react to the environment—this is personality.  It could be a theory about the organization’s business model and how employees contribute to the model—this is performance.  It could also be a theory about how people should relate to each other and themselves to support the organization—this is a competency model. These theories and models are all helpful tools for understanding and describing human capabilities and outcomes.

The Strengths of Mental Models

Models are helpful. Architects, boat builders and other craftspeople have used them for years. In a management context, we need mental models of talent to understand employees, to know how employees contribute to the larger operation, and to be able to predict how employees will react in a range of situations.  Without these powerful tools, we could not effectively manage our talent. 

Competency models work so well because they make these mental models explicit and transparent, and because they allow us to articulate the behaviors that are related to performance. Explicit competency models have radically changed how talent is managed. In the past, a manager might have said only that an employee needed to be a better team player. A competency model gives the manager an elaborate description of what it means to be a team player, and describes the behaviors in terms that can be communicated, measured, and emulated. 

When these behaviors are measured, competency models support better insights, more motivation, and obvious decisions. For example, the Danielson framework is a model of teacher performance that allows organizations to select, train, coach, and improve educator performance using a single set of expectations.

Competency models measure observable behaviors; personality assessments, which describe innate natural tendencies, offer another set of powerful tools. A coach who has a strong understanding of a personality system (for example, the MBTI or the HDI) can assess someone to gain insights and then coach using the framework.  A manager who has a strong mental model of personality is better able to see consistencies and predict how others will react to situations.

The Problems with Mental Models

Although models of personality, performance, and competency are powerful places to start, we often forget that any model is a simplification of a complex reality. A personality measure focuses only on a few aspects of an individual’s nature; a performance measure considers only one contribution to a business; and a competency assessment considers only a few human capabilities.

 There is also a danger when we’re not aware that we’re using a model. Mental models, as defined in organizational system dynamics, are deeply held images of thinking and acting. Mental models are so basic to understanding the world that people are hardly conscious of them, and this leads to problems. 

For example, if I’m talking with an employee about her performance, we may be talking about two different things. My employee may be focused on the quality of her writing and communication, while I’m focused on the number of billable hours. We’re talking about doing a great job—and we’re completely miscommunicating. Both of our models are necessary simplifications. One is necessary from a business standpoint, and the other from the standpoint of doing the work.

As this example shows, when our mental models are implicit—not apparent to either person—they limit our perceptions and prevent us from deliberately acting and communicating.

Implicit mental models of talent lead to misalignment, miscommunication, and narrowed focus.

Miscommunication. Good communication is based on shared meaning. Words like intelligence, personality, or performance mean different things to different organizational stakeholders. Unacknowledged mental models of these critical talent constructs lead to miscommunication.  We may be talking, but if our meanings are different, we are not communicating.

Narrowed focus. Mental models provide a framework for what we should pay attention to. The problem is that in looking for one aspect of personality, performance, or competency, we may miss another, equally important factor. 

For example, a personality model directs our attention to behaviors that suggests a personal tendency to react in a predictable way. We may miss other behaviors that would tell us something else about how the individual can contribute. I have colleagues who have such a strong understanding of the DISC personality system that they immediately notice that someone is largely Dominant, Influence, Steady, or Compliant. They are so good at classifying others using this system that they miss other aspects of their personality.

This is a shame, because there are many ways to look at personality.  The simplest model has four factors, but more complicated models exist, including the 16PF or the Caliper Profile.  A more complicated model allows for more refined insights.

Most aspects of talent are multidimensional. A person may be high in Dominance, but nearly as high in another dimension. Further, different aspects of personality may appear in different situations.  My colleagues’ mental models may be limiting their expectations of others to something much more simplistic than it actually is.

Misalignment. To be useful, talent models must align with organizational needs.  If we are unaware of our models, this may not be the case.

Think back to how life insurance sales representatives are paid.  If you’re not familiar with the insurance 

industry, it seems odd.  However, it is perfectly sensible to an insurance insider. One of the strengths of the industry is the stability associated with customers paying premiums year after year for their entire life, and only collecting a payout when they die.  Because of this, long-term relationships and accountability is important.  In this sense, the model of sales representative pay is aligned with corporate strategy. 

As organizations change, our mental models of talent must also change. Unlike a desk or a rock, talent can change and adapt.  New insights and technologies can suggest to better, different, or more detailed models of talent.  Often, this is an opportunity for growth and development.

However, if we are unaware of our mental models, they are difficult to change. 

For example, salespeople in an organization moving to a team-based sales environment must be able to examine their assumptions, and must be aware of their own mental models, because the organization is changing the model. Performance is no longer individual. The change will affect the team, its management, and the support of the team. Team members will have to change how they view performance, the management will have to think differently, and the measurement and pay systems will have to change.

Thinking Differently

Teachers today face more competition for children’s minds.  It may be that engagement is more important in an era of video-games and 24/7 entertainment, so the new model of teacher performance is appropriate for today.  However, the education system has a difficult task in getting veteran educators to think differently.  For too long, assumptions about effective teaching were based on outdated thinking. Worse, we were not aware of our assumptions about teaching.

If we are unaware of our model of effective teaching, we will have a hard time discussing change, let alone changing. If we are not aware of the model we are using, we will not manage talent optimally.  

The bad news is that when we measure talent, we always are measuring a model. The process of developing measures that represent the model we tend to made something seem extremely concrete.  The model is transformed into something more real than it actually is.  

The good news is that we have become much more sophisticated in our thinking about talent models. The competency revolution made behavioral models apparent. Now we simply have to remember that underlying every measure is a mental model of talent.  Remembering this will help us question our assumptions, articulate our mental models, and test alignment with organizational direction.

In the next blog post we will consider reification of talent measures.