In their Agility Factor article, Thomas Williams (Booz & Company) and Christopher G. Worley and Edward E. Lawler III (both of the Center for Effective Organizations, University of Southern California), defined ‘agility’ as:

a cultivated capability that enables an organization to respond in a timely, effective, and sustainable way when changing circumstances require it.

They identified four ‘routines of agility’ that differentiated the high performing organizations: strategizing dynamically, perceiving environmental change, testing responses, and implementing change as outlined in their Exhibit 2 below and discussed in my previous post.

How do these agility routines relate to Adaptive Iteration?  I believe that Adaptive Iteration is a way of building and embedding the organizational capabilities and individual expertise necessary to support the agility routines as follows:

Strategizing dynamically

Sense of shared purpose: is incorporated as a key element of Adaptive Iteration.  A clear and shared overarching purpose is critical to guide the many Adaptive Iteration decisions and choices.  It also provides the high-level scope necessary to enable and stimulate creative and innovative improvement options.

A change-friendly identity: is encouraged and stimulated by Adaptive Iteration.  The Observation and Interpretation capabilities embedded in Adaptive Iteration encourage and develop a realistic assessment of the alignment of the organization with its purpose and operating environment.  The resultant shared recognition of any misalignment creates the collective tension and energy necessary for constructive change.

Robust strategic intent: is an outcome of applying Adaptive Iteration to the organization’s business strategy.  Adaptive improvement approaches, such as Adaptive Iteration, excel in the emergent and partially predictable context of most business strategies.  Adaptive Iteration is a fractal capability that can be used not only to change and improve a business strategy but also to stimulate and guide the necessary changes to the organization’s capabilities and resources.

Perceiving Environmental Change

Sensing: corresponds with the Observe phase of Adaptive Iteration.  Adaptive Iteration requires, encourages and develops incisive and unbiased observation of all aspects of the organisation and its environment.  It provides a framework within which observations (sensing) can trigger adaptive responses.

Communicating: is embedded in the iterative nature of Adaptive Iteration; in the emphasis on multiple perspectives during the Observe phase; in the emphasis on diversity during all phases, but especially Interpret and Design; and in the progressively expanding nature of the Experiment phase.  Open communication is embedded in Adaptive Iteration and should grow throughout the organization as a result of its use.

Interpreting: corresponds with the Interpret phase of Adaptive Iteration.  Adaptive Iteration recognizes that many situations are an ordered/random hybrid that cannot be analysed with precision or predicted with certainty.  However, interpretation of observations can identify important change-related characteristics such as early signs of failure or success, non-linear patterns, key constraints and decision points, and driving influences.

Testing Responses

Slack resources: are made available when adaptive improvement (ie Adaptive Iteration) is recognised and acknowledged as a necessary and routine organizational capability. Embedding a framework such as Adaptive Iteration creates the structures, capabilities and triggers within which ‘slack resources’ can be released and applied.  It bridges to the existing discipline and rigor required for operational excellence.

Risk management and learning: are embedded in the iterative nature and emphasis of Adaptive Iteration and in its requirement that all iterations be guided by a clear and commonly understood overarching purpose.  The explicit separation of observation, interpretation, design and experimentation reduces the potential influence of biases and predetermined expectations.  Adaptive Iteration and its enabling capabilities promote and encourage purposeful learning-based action throughout the organization.

Implementing Change

Management and organizational autonomy: is encouraged and facilitated by the structured and rigorous nature of Adaptive Iteration.  Adaptive Iteration is not an ad-hoc or occasional approach to developing and implementing change and improvement.  It does not require close management involvement and oversight.  Rather, as with the organization’s operational processes, it is a disciplined processes supported by well-defined capabilities and routines and governed by the organization’s overall purpose.

Embedded change capability: occurs when Adaptive Iteration (or any other structured approach to adaptive improvement) is adopted throughout the organization.  The clear discipline and structure of Adaptive Iteration is a natural extension of the organization’s operations.

Performance management: arises from the use of an overarching purpose or intent to drive decisions and choices throughout Adaptive Iteration.  Care must be taken to ensure that narrowly set performance objectives and measures do not limit creative or innovative options and do not drive the iterations to a premature and sub-optimal conclusion.

In summary, the work of Williams, Worley and Lawler has highlighted a correlation between organizational agility (as assessed by four agility routines) and company profitability.  Adaptive Iteration is an approach to developing organization agility and in this post I have explored how it maps to the four agility routines described by Williams, Worley and Lawler.

Photo: Derek Jensen – Wikimedia commons

The Strategy+Business web site recently reported on an in-depth study that looked at the relationship between agility and long-term business performance.  The authors, Thomas Williams (Booz & Company) and Christopher G. Worley and Edward E. Lawler III (both of the Center for Effective Organizations, University of Southern California), report that:

When the measure of performance is profitability, a few large companies in every industry consistently outperform their peers over extended periods. And they maintain this performance edge even in the face of significant business change in their competitive environments. The one factor they seem to have in common is agility. They adapt to business change more quickly and reliably than their competitors; they have found a way to turn as quickly as speedboats when necessary.

The study looked at the Return on Assets (ROA) for 243 large firms in 17 industries over the 30 year period from 1979 to 2009 and identified the percentage of years for which the firm performed above the industry average.  Return on Assets was compared with survey results that evaluated each company against each of four ‘agility-related routines’: strategizing dynamically, perceiving environmental change, testing responses, and implementing change.  As with the ROA scores, each ‘agility-related routine’ result was assessed to be above or below the average of all companies in the survey.  The following chart summarizes the results.

The study results indicate that above average agility performance generally correlates with above average financial performance in terms of Return on Assets.  Further, the results reinforce that agility is a systemic capability in organizations.  The companies in the study needed above average performance in at least three of the four ‘agility routines’ to achieve consistent above average financial performance.  This finding is consistent with my conviction that organizational agility and adaptive capability requires Adaptive Iteration – a purpose-driven iterative cycle of observation, interpretation, design and experimentation.  More on that in my next blog post.

As mentioned in the previous blog, I have recently finished the draft of a comparative review of eight existing approaches that are similar to Adaptive Iteration.  These eight approaches are listed in the table below, together with a listing of how each approach describes the the four phases of adaptive iteration.   As can be seen from the table, there is considerable similarity between the eight approaches, but the review also identified a range of subtle but significant variations, most of which are due to the context for which each approach was first developed.  A brief summary of each of eight approaches is provided below.  A more detailed summary, together with some comparative analysis is provided in the White Paper.

Developer	Title	Phase of Iteration
		Observe/ Sense	Interpret/Learn	Plan/Decide	Act
Kolb	Experiential Learning	Reflective Observation	Abstract Conceptualiz-ation	Active Experimentation	Concrete Experience
Boyd	OODA Cycle	Observe	Orient	Decide (Hypothesize)	Act           (Test)
Deming/ Shewhart	PDCA/PDSA Cycle	Check/Study (Observe)	Act       (Learn)	Plan	Do          (Test)
Haeckel	Adaptive Loop	Sense	Interpret	Decide	Act
Dixon	Organizational Learning Cycle	Integrate	Interpret	Act	Generate
Snowden	Cynefin: Complex Domain	Sense		Respond	Probe
Ries	Lean Startup	Measure	Learn	Build
Eoyang	Adaptive Action	What?	So What?		Now What?
Fietz	Adaptive Iteration	Observe	Interpret	Design	Experiment

Table 1.  Variants of the Adaptive Iteration approach

Kolb’s Experiential Learning model emphasizes concrete experience (action) as the source of learning.  Kolb highlights that Experiential Learning requires that the learner have four kinds of abilities – concrete experience, reflective observation, abstract conceptualization, and active experimentation. The effectiveness of Experiential Learning depends on the extent to which the learner can apply and integrate all four abilities to an experience.  Kolb believes that the ability to resolve and integrate the inherent conflicts between the four “adaptive modes” (abilities) is a “hallmark of true creativity and growth”.

The OODA (Observe-Orient-Decide-Act) loop was developed by John Boyd, a fighter pilot in the US Air Force, to explain why U.S. pilots flying F-86s fared so well in air-to-air combat against enemy MiGs during the Korean War.  Boyd subsequently developed a reputation as a  military strategist and ‘thinker’.   The simplicity of the OODA Loop model led to some criticism of its suitability for more general use (outside combat situations) and this led Boyd to propose a more elaborate version of the model.  In particular, the elaborated version included more feedback paths and provided for contextual considerations, especially for the Orient step.

The PDSA/PDCA (Plan-Do-Study-Act/Plan-Do-Check-Act) Cycle was developed in an early form by Walter Shewhart and refined and enhanced by W. Edwards Deming.  It was developed primarily as a participative approach to the continuous monitoring and improvement of manufacturing (production) processes.  It is one of the core methods of the Total Quality / Continuous Improvement approaches developed and refined in Japan through the 1950s, 1960s and 1970s and subsequently adopted widely by manufacturers in western countries.  Deming continued to refine and adapt the ‘Shewhart Cycle’, as he called it, and the terminology in the version published in 1986 is more applicable to less structured and predictable processes:  ‘Observe’ replaces ‘Check/Study’, ‘Learn’ replaces ‘Act’, and ‘Test’ replaces ‘Do’.

Stephan Haeckel incorporates his Adaptive Loop into his design for Sense-and-Respond organizations.  Haeckel’s Sense-and-Respond organizations are those that deliberately pursue the ability to dynamically sense market needs and rapidly organize to fulfil them.  This is in contrast with the more conventional Predict-and-Plan organizations which, because they are not structured for rapid response to market changes, place greater emphasis on predicting the market and then using those predictions as the basis for future plans.  The Adaptive Loop is less of a loop and more a repeated application of a linear Sense-Interpret-Decide-Act process.  It is distinguished by the precision and short cycle time of that end to end process.

Nancy Dixon uses an iterative model, called the Organizational Learning Cycle, to describe the requirements for organizational learning.  Not surprisingly given Dixon’s focus on learning rather than action, her emphasis tends be on people-related structures and processes.  The emphasis is on creating shared and contextually relevant meaning through widespread collection, sharing and interpretation of information.  As with Haeckel, there is limited emphasis on the creation of feedback based learning.

In their Cynefin sensemaking framework, David Snowden and Mary Boone propose a Probe-Sense-Respond approach as one of a suite of leadership actions in the Complex domain.  The Cynefin Complex domain corresponds with the domain of interest in this review, that is, domains that are partially structured, emergent and partially predictable.  Snowden emphasizes that the Probe phase consists primarily of safe-to-fail experiments that both test a hypothesis about potential improvements and generate additional learning about the nature of the Complex System.  He also allows the possibility of multiple parallel experiments to increase the scope and depth of the learning.  Snowden also advocates the use oblique or indirect approaches to identifying improvements and encourages the introduction of diverse perspectives by incorporating related-but-different expertise.

According to Wikipedia: “Lean Startup” is an approach for launching businesses and products, that relies on validated learning, scientific experimentation, and iterative product releases to shorten product development cycles, measure progress, and gain valuable customer feedback.  The Lean Startup approach began as a methodology for (software based) entrepreneurial startup activities.  However, it is increasingly being applied within large companies to address situations with that have significant levels of uncertainty – a defining characteristic of the Lean Startup approach.

Glenda Eoyang and Royce Holladay of the Human Systems Design Institute use a three phase iterative ‘Adaptive Action’ approach to address uncertain and unpredictable situations.  The approach draws heavily on the concepts and perspectives of Complexity Science, particularly those relating to the self-organizing behaviour of Complex Adaptive Systems.  Central to the Adaptive Action approach is the Containers, Differences and Exchanges (CDE) construct.  The language and case examples in Adaptive Action: Leveraging Uncertainty in Your Organization focus primarily on organizational development, organizational behaviour and leadership.  No doubt, the general approach has wider application, but it is unclear whether the current formulation and language of Adaptive Action will stimulate broad application outside the organizational/leadership context.  That remains to be seen.

We propose Adaptive Iteration as a framework to develop the core capabilities necessary to develop agility and adaptive capacity in organizations.  Adaptive Iteration emphasizes the co-evolution of learning and action and the importance of a clear and shared purpose to guide the adaptive process.  It draws on the lessons and approaches of those who deal with unpredictability and uncertainty every day – entrepreneurs, designers, researchers and developers.

I have just completed a Discussion Draft of comparative review of eight existing approaches that are similar to Adaptive Iteration.  I have published the comparative review as a White Paper here.  One of the benefits the review is that it has helped clarify and crystalize the essence of Adaptive Iteration in the form of a set of principles – still in draft at this stage.  I listed the principles towards the end of the White Paper and repeat them here for you to review and comment.

Principles for Adaptive Iteration

1. Adaptive Iteration applies to all non-random situations where behaviour and performance cannot be predicted with confidence in advance.  In the main, this applies to situations that are partially structured and emergent.  Because of dramatically increased connectivity generated by developments such as globalization and by technologies such as the internet, situations that require Adaptive Iteration are increasingly prevalent and of increasing significance.

 

2. Adaptive Iteration involves the co-evolution of learning and action in an iterative four phase approach:
   • Clear and unbiased observation of the situation, including contextual influences
   • Interpretation of the observed data to improve understanding of the trajectory of and the influences on the situation
   • Incorporation of the interpretation insights into a hypothesis for an improved design
   • Conduct of one or more experiments to assess the impact of the proposed improved design.
3. Adaptive Iteration is underpinned by a clear and shared understanding of the fundamental drivers for and constraints on the adaptive decisions.  This provides coherence and alignment for all decisions and provides lower level scope for the adaptive process to explore novel and unconventional options.  It also provides a clear driving force for adaptation and against the status quo.

 

4. Adaptive Iteration may be a nested approach in which a lower level Adaptive Iteration cycle is used as the experimental phase for a proposal for a higher level design change.  Because of its multi-level applicability, Adaptive Iteration should be a core capability throughout any organization.

 

5. Rapid iterations of Adaptive Iteration are important during the early stages of an adaptive transition.  These iterations promote critical early learning and integration across the four phases.  They also help shake-out any lack of shared clarity about the fundamental drivers and constraints.

 

6. Adaptive Iteration co-exists with analytical action.  Where the impact of a design variable can be predicted efficiently and with confidence, Adaptive Iteration is not required and the design changes should be subject to ‘analytical action’.

At Discerning Action we believe that Adaptive Iteration is a valuable approach to working with unpredictability and uncertainty in organisations.  What do we mean by Adaptive Iteration?  What are the core elements and how do they fit together?

For the purposes of this overview I will begin with the Design step in the Adaptive Iteration cycle (although that may not always be the case in practice).

A design is a hypothesis about the most appropriate response to a complex and unpredictable problem or opportunity.  The design hypothesis could relate to a physical product, a service, a plan of action, a strategy, a business model, an organisational development initiative, etc.  A design is a hypothesis because until it has been successfully tested in context through one or more experiments, its suitability as a response is not proven.

Experiments can take various forms, including a thought experiment, a prototype, a simulation, a trial or a pilot.  A unique natural occurrence in a business or organisational context could also be considered a form of unplanned or natural experiment.  In such a case, the Adaptive Iteration cycle would start with the Experiment step.

An experiment could be predominantly confirmatory or predominantly exploratory. In the latter case, the primary objective would be to stimulate information or insights (learning) on which the next iteration of design is based.  I believe the expertise to formulate and execute various types of experiments is a critical aspect of Adaptive Iteration, and is not well developed in many organisations.

An experiment is of little value unless it is supported by accurate observation.  However, the observer’s challenge is that in observation our brains are heavily influenced by what we expect to see.  Our observations are significantly biased by our, often subconscious, experience and expectations.  If the context for our observations is stable and predictable, this bias provides us with significant cognitive processing advantage by making it fast and efficient.  But it is a major weakness if the context is complex and unpredictable.  In such cases, the Observe step requires explicit consideration and the development of specific observational expertise.  As Louis Pasteur once said, “in the field of observation, chance favours the prepared mind”.  When observing in complex and unpredictable contexts we need to “prepare our mind”.

Experimental observations are useful only if they lead to relevant learning through interpretation.  The learning may relate directly to the initial design hypothesis, or to unexpected observations that arose because they were made with an open mind and/or from multiple perspectives.

In complex and unpredictable contexts the types of things we seek to interpret from our observations include: critical decision points and options; weak signals that may indicate an emerging coherence; the early stages of a reinforcing feedback loop that may trigger rapid change; indicators of the underlying driving forces and motivators in the system; and the nature and impact of the constraints and boundaries of the system.

Based on the results of the experiment, the design will be further modified or refined.  These changes are, in effect, another hypothesis to be tested and evaluated by further Adaptive Iteration.

At the center of Adaptive Iteration is a clear and shared understanding of overall intent or purpose.  This creates coherence for the myriad of decisions that must be made when undertaking Adaptive Iteration.  A shared understanding of overall Purpose is especially important given the complex and unpredictable nature of the context.  The Purpose should be at a reasonably high level otherwise it will constrain the nature and scope of the design hypotheses that are generated.  If it is at too low a level, it will already have significant design decisions built in and will reduce adaptive scope.

 

Many of the issues and trends facing organisations today are complex and, to a large degree, unpredictable in terms of their specific impact and timing.  They include climate change, globalisation, social media, big data, internet of things, rise of China and other industrialising countries, global financial crisis fallout, and the retirement of the baby boomer generation.  These issues and trends offer opportunities for many but also have the potential to disrupt existing business and organisational models through rapid and unpredictable change.  As a result, agility and adaptability are, and will increasingly be, critical organisational capabilities.

However, the current dominant approach for solving problems and realising opportunities in organisations is based on the underlying assumption that the context for the problem or opportunity is well described and amenable to analysis and prediction. This assumption, in turn, leads to an essentially linear ‘analyse-plan-implement’ approach to creating or responding to change.  In addition, this approach underpins almost all teaching in professions such as engineering and commerce.  By itself, the ‘analyse-plan-implement’ approach will not be sufficient to address the emergent and unpredictable issues and trends typified by the list above.  Nor will it be sufficient to create the level of agility and adaptive capability necessary to sustain organisational health and success.

The ‘analyse-plan-implement’ approach needs to be supplemented by capabilities, methods and techniques that iteratively create solutions and responses based on a Purpose-driven adaptive cycle of Design, Experimentation, Observation and Interpretation (which I will refer to as Adaptive Iteration).

Our purpose at Discerning Action is to develop and demonstrate Adaptive Iteration as a practical approach for building agility and adaptive capacity in organisations, and to present techniques for and examples of its application.