Opinion Piece: Rethinking Performance: Why Behaviour Remains The Missing Link In Evaluation

Rethinking Performance - Part 2

Part 1 argued that performance systems are incomplete when they privilege outcomes over the conditions through which those outcomes are created. In this second article, written by Camille Rabier from https://www.21century.co.za/, we explore why behaviour is so difficult to evaluate in practice, and why that difficulty sits at the heart of the design problem.

This second article explains why behaviour is so difficult to evaluate in practice, and why that difficulty sits at the heart of the design problem.

Why performance systems struggle to evaluate behaviour

Behaviour is critical to performance, yet it is rarely reflected in how performance is assessed. This is not simply a design oversight; it reflects the inherent difficulty of evaluating behaviour. Unlike outcomes, behaviour is not easily defined, observed, or interpreted consistently. Understanding why requires examining the nature of behaviour itself.

Behaviour is context-dependent

The same behaviour does not have the same impact in every situation. A behaviour that is effective in one context may be ineffective, or even detrimental, in another. Some behaviours may deliver short-term results while weakening the organisation over time, whereas others may sustain performance or create the conditions for future performance.

Behaviour can also create or constrain other behaviours. For example, working independently may drive efficiency and speed in some contexts, but in others it may limit collaboration, knowledge-sharing, and broader organisational alignment. As a result, the effectiveness of behaviour cannot be assessed in isolation. It depends on the context in which it occurs, the conditions it creates, and the impact it has over time.

What constitutes 'good' behaviour is therefore not fixed, but contingent on the value it creates and whether that value is sustainable.

Behaviour is interpretive

Because behaviour is inherently human and social, it is not simply observed but interpreted. How behaviour is perceived and evaluated is shaped by individual judgement, experience, and cognitive bias. As a result, different managers may respond to the same behaviour in different ways, forming different conclusions about its impact. These interpretations are influenced not only by what is observed, but by how individuals are predisposed to perceive and evaluate behaviour. Observation therefore does not produce a single objective assessment, but a range of interpretations.

This makes it even more difficult to establish consistent standards, as the same behaviour may be evaluated differently across individuals, contexts, and situations, leading to different conclusions about its effectiveness and contribution to performance.

Behaviour is multi-dimensional

What is often described as a single behaviour is, in practice, made up of multiple underlying elements. Behaviours such as ownership, collaboration, or accountability are not singular actions, but combinations of different capabilities, choices, and actions expressed over time. For example, collaboration may include communication, information-sharing, alignment, responsiveness, and collective decision-making. As a result, behaviour is difficult to define cleanly because it does not exist as a single, discrete unit that can be easily isolated, measured, or compared.

Evaluating behaviour therefore requires interpreting a combination of signals, rather than assessing a single observable indicator. This introduces a further complexity: if behaviour is made up of multiple elements, do they all carry equal weight? Not necessarily. The relative importance of each element can shift depending on the context, the role, and the outcome being pursued. For example, strong decision-making may be more critical than collaboration in one situation, while in another, the opposite may be true.

This makes it even more difficult to establish consistent evaluation criteria, as assessment requires not only identifying the relevant elements, but also determining how much each should count.

Behaviour is continuous

In addition to being composed of multiple elements, behaviour is also expressed over time rather than at a single point. Unlike outcomes, which can often be tied to specific events or points in time, behaviour is expressed continuously. It unfolds over time through patterns of action rather than single, isolated instances. As a result, behaviour cannot be captured through once-off observations, and a single moment is rarely sufficient to represent how someone typically operates on a day-to-day basis. Evaluating behaviour therefore requires forming a view over time rather than assessing a discrete event, which creates a challenge for performance systems that are typically structured around fixed review periods and documented evidence.

Because behaviour does not naturally align with these boundaries, it is harder to observe consistently, evidence reliably, and compare across individuals, as assessments may be based on partial or uneven observations.

Behaviour is entangled with outcomes

Ultimately, outcomes are shaped in part through behaviour. They do not occur independently but are influenced by how work is carried out. As such, no outcome can be meaningfully interpreted without considering the behaviours that contributed to it. The two are inherently intertwined. For example:

• Strong outcomes can be achieved through behaviours that are unsustainable or detrimental.
• Effective behaviours may not always produce immediate or visible results.

As such, the very same outcome may reflect very different underlying behaviours, and similar behaviours may lead to different outcomes depending on context, timing, and external factors. This means that outcomes cannot reliably be used to infer underlying behaviours. In practice, however, performance systems often default to outcomes in the absence of clear behavioural criteria, creating the impression that performance has been fully understood through the impact delivered while overlooking the underlying value created through behaviour.

Behaviour and outcomes are therefore not interchangeable, nor can one compensate for the absence of the other. Performance can be understood only through the relationship between what is delivered and how it is achieved.

Implications for performance systems

The difficulty in evaluating behaviour is therefore not the result of poor design, but a consequence of the nature of behaviour itself. Its inherent qualities make it difficult to define, observe, and translate into consistently applicable criteria within performance systems. This stands in contrast to the requirements of performance systems, which depend on clarity, consistency, and comparability. As a result, performance systems tend to prioritise what can be measured reliably - namely outcomes - while behavioural criteria remain under-specified and inconsistently applied.

In practice, this leads to partial representations of performance, where what is most visible and defensible is assessed, but not necessarily what is most valuable or sustainable.

The challenge, therefore, lies in how behaviour can be translated into something usable within performance systems.

Ends.

Total Words: 1047