Rethinking Performance Management: Beyond the Linear Approach 

Introduction

We are approaching the sacred three months of performance management, which is quite famous in the corporate world. To be honest, the thing I disliked the most when I was a manager was performance management, especially the process of determining performance ratings within a certain quota at the end of every year. Therefore, please be forewarned if there is a whiff of prejudice in this article.  

The Challenges of Performance Management

The foundations of performance management, or perhaps in different terminology, Management by Objectives, can be traced back to Taylor's scientific approach to management more than 100 years ago, which laid the foundations of industrial engineering and perhaps all management approaches today, with the aim of increasing the productivity of workers and work. He was pretty successful in establishing a system that would determine how much work could be done in a given time through time studies and systematic analysis, setting performance targets based upon them, and financially rewarding employees who met these targets. All performance management systems today seem to be trying to do more or less the same thing - don't they?   

Only we have a few extra problems. The first one is that Taylor's approach can be used in industrial enterprises where production systems are dominant and where most of the work is repetitive and done by machines because it is not always easy to do time studies or systematic analyses of tasks for a knowledge worker. Believe me, I tried; I tried hard in a product development environment and failed miserably. The second challenge of today's enterpriser is the acceleration of change and competition in the information age: we need faster goal updates and feedback loops.  

The Evolution of Performance Management: From MBO to OKRs

The Evolution of Performance Management from MBO to OKRs

Considering that the focus of the famous management guru Drucker was knowledge workers, it seems that he wanted to overcome this challenge. In the 1950s, he introduced the idea of Management by Objectives (MBO), suggesting setting goals that are aligned with the company's objectives and evaluating employee performance based on these goals.  

Later on, the balanced scorecard emerged, separating financial and non-financial results. I don't know if there was a very obvious difference from MBO, but I started my career during the period when this balanced scorecard was being implemented, and I graduated from my corporate life with OKRs. OKR system has many good applications compared to the Balanced Scorecard; it is more refined, more agile, and more transparent. I later learned that this was also not a very new invention. It was invented at Intel in the 1970s. In fact, the main logic of the OKR approach is to create unity and focus in the company, as well as to set and achieve goals. As I experienced it, apart from its main purpose, OKRs were directly used as the main input for annual performance evaluation and have become the foundation of performance management.  

Now, enough with the background information.  

In fact, it seems to me - for knowledge work - that the main purpose of this whole idea of 'management by objectives' is to establish a closed-loop control system in the work systems of knowledge workers. Let's salute Henri Fayol here because he said in 1916 that 'management has five activities: planning, organisation, command, coordination, control.

What did he say?

Planning and control.

Looking back on all this management history, we see that we want to plan and control things. But is this bad? Probably not, but it depends on what we want to control.  

In fact, the idea of closed-loop feedback and control systems is not something unknown to us; in fact, it is all around us. The simplest example is the flushing system in toilets.  

When we flush the toilet, a float system inside the bowl gives a signal, that is, it opens the valve and the water starts to flow into the tank then when the water reaches a sufficient level and lifts the arm of the float, it gives another signal, that is, it closes the valve and the water stops flowing. It is literally a 'self-regulating' system. The scientific name for this is cybernetics. Of course, I understand that it doesn't sound scientific when I explain it through the toilet flush, but in fact, almost all of today's management approaches can be said to be based on this cybernetic control. When you look at it more carefully, what Scrum is trying to do is to make it a more systematic iterative cycle in a similar way to emphasise experimentation to find solutions to complex problems. But let's avoid going in there and over-complicating things.  

What we were saying was that the cycle of management by objectives and even performance management is actually an attempt to turn the productivity of us knowledge workers - who have the means of production between our ears - from an open system to a closed loop system, and thus to plan and control it, because our productivity is not predictable like in machines.  

To set up closed-loop cybernetic control, you need an input signal. These are the goals that are given at the beginning of the year: the plan. There needs to be a feedback signal, and that is the end-of-the-year performance evaluation. A target will be given, where we are in relation to that target will be measured, feedback will be given, and employees who observe the difference between target and performance will be expected to regulate their productivity just like a flushing system regulates water. Control.

It makes a lot of sense when you think about it, but this is exactly the product of reductionist thinking. The relationship between the valve controlling the flushing system and the water flow can be solved by very simple equations with few variables, so reductionism in flushes is very practical for us. This is the case in many areas of our lives. The operation of air conditioners, cruise control in cars, etc., all give predictable results with the help of cybernetics, i.e. closed-loop feedback, control systems and mathematics.

However, the productivity of people and knowledge workers is not a linear equation with few variables. It is a multivariable, non-linear situation. What I am trying to say is that the float in the siphon depends on the laws of physics; how it will move can be mathematically calculated and predicted, and rational causality applies, but people and their behaviour cannot be mathematically predicted; their behaviour depends on many variables such as their motivation, cognitive state, etc., rational causality cannot always be expected.  

Moreover, the relationship between the results of individuals' performance evaluations and business results is entirely context-dependent. The financial success of a company depends as much or even more on the economic conditions of the country in which it is located as on the influence of its CEO. Since we directly link business results, OKRs and targets to performance evaluations, almost everyone leaves this process unhappy. I once came across a team member who had received an 'A' but was unhappy because he had received an 'A' based on business results, but a teammate who contributed to those business results at the same rate had received a C due to quotas. This friend's sense of justice was affected.  

I think the reason why performance management tools never work is that we try to solve a multivariate non-linear problem (complex) with a linear and reductionist approach.  

In my opinion, almost 90% of the time managers spend on performance appraisals is wasted and produces undesirable results.  

Closing Statements

So what should we do - should we not do performance management, or should we leave the system open-ended? I would say NO.   

I don't have a clear prescription (and complex problems cannot be solved with prescriptions), but I may have some suggestions:  

1. My most important suggestion is to separate management by objectives or your OKR system from an individual's performance management system. Even though we have a strong attachment to controlling people, let's part with love from that attachment. Manage the business with objectives or OKRs; evaluate employee performance with another method. In other words, Manage the System, Not the People.
2. Respond to complexity with complexity.
3. The brain is more complex than any other tool. Give space for people to do performance management, trust them, and leave it to teams.
4. Be open to different views.
5. Improve your performance management tools day by day with experimentation and short feedback loops.