Benchmarking typically measures performance and outcomes on the basis of best praxis observed by others or by oneself in the past.
Many systems thinkers (e.g. Ackhoff) regard benchmarking per se as unsystemic. A key argument is the mediocrity produced by comparing with and following others versus the creativity of being a leader and innovator (i.e. the uniqueness that arises if one applies systems thinking).
By analogy, one cannot meaningfully benchmark Picasso with Rembrandt, or Madonna with Mozart, but we probably could benchmark the disciples of each master and benchmark them on how well they copy / imitate him or her.
Having made a general statement, I would nevertheless like to consider the issue of benchmarking in general, and of systems thinking specifically, in more detail. I would like to start by reflecting on it from the perspective of different systems concepts to determine if and in what context benchmarking of systems thinking is useful or not. (See also www.biomatrixtheory.com for further explanation of these concepts).
One of the core concepts of systems thinking is co-production. Applied to benchmarking, this implies that one cannot isolate one co-factor, like practicing systems thinking, and make it responsible for the (successful or unsuccessful) behaviour and outcomes of an organisation. Other co-factors could be much more important (e.g. new opportunities in the environment could be responsible for success, while the best praxis of systems thinking could fail to deliver good results because of other co-factors).
Another concept relevant to benchmarking is that of emergence, namely the emergence of new qualities and outcomes from the interaction of a system with other systems in its environment. Describing the characteristics and outcomes of a system involves two types of qualities: type I and II.
Type I qualities are inherent in the system and tend to lend themselves to measurement (e.g. number and amount of resource use, efficiency measures, speed of processing, waste and faults produced, etc.). They are typically the subject of benchmarking and benefit by it.
Type II properties are properties that emerge from the interaction of systems and are not inherent in one of the interacting systems. These qualities tend to be unique to a specific interaction, not repeatable (at least not in exactly the same way) and not measurable. Examples are the evaluation of innovation, strategy development, ideal designs and new product development.
The aim of introducing systems thinking in an organisation can enhance both types of qualities. Type I qualities can be improved through applying systems thinking in an operational context (e.g. through systems optimization which is aimed at producing more efficiencies). Applying systems thinking in a strategic context (e.g. for business and organisation development) should produce unique and creative characteristics and outcomes (type II) which do not lend themselves easily (if at all) to benchmarking.
current versus ideal logic
Comparison implies current logic (i.e. repeating more of the same thinking) that leads to the replication of the functioning of the system that serves as a benchmark. By comparison, ongoing development requires creative behaviour of the system within its continuously changing context and driven by its unique aims (e.g. its unique vision of an ideal future).
Benchmarking the ideal (even in its current approximation) does not make much sense. For example, the unique interpretation of a vision like wanting to be the most beautiful restaurant in town, or leaping ahead through technology, cannot be benchmarked. However, once this interpretation has been manifested (e.g. as a particular colour scheme, furnishing and service of the restaurant, or a new feature in the car), it can be compared and measured.
But how much of all this is due to having entrenched systems thinking in the organisation? Good strategic thinking and various creativity techniques taught in every MBA course could lead to similar outcomes.
A similar argument can be derived from the observation that systems embrace paradox. Systems are conservative as well as progressive. Due to their structure, they are relatively stable, behave routinely and operate in a standardised manner, producing standardised outcomes (i.e. type I qualities). Adapting to a continually changing environment, they are also flexible, creative and synergistic (i.e. they display type II qualities).
To successfully benchmark, one would need to distinguish between different parts and behaviours of the system – those which serve standardisation and can be benchmarked and those which require creativity and cannot.
different types of governance
From a governance point of view, entrenching benchmarking is form maintaining governance, not a form creating one as would be required for new development. The outcomes of form creating governance (e.g. creative strategies, new product development) are unique and do not lend themselves to benchmarking. The outcomes of form maintaining governance do. This could refer to any regular activities associated with or designed by a systemic intervention. For example, the operationalisation of systems thinking could be measured through the regularity of planning meetings that are held after the systemic redesign of the organisation. Or one could measure if the planning meetings use systemic methods (e.g. stakeholder analysis, systemic brainstorming). But the appropriateness of their use or the outcomes they produce cannot be measured.