Kevin Eyre, of SA Partners, shares an interesting analysis and comparison of lean, appreciative inquiry and work-out and urges practitioners to explore the plethora of different methodologies available out there.

This article is a celebration of the diversity and natural creativity of people determined to improve the organisations in which they work. It is also a ‘call to arms’ for leaders of change to avoid the simplistic adoption of standard approaches to business improvement and to think and experiment their way to the creation of change and improvement strategies that work for them. But first, the celebration.


Of the many approaches to organisational improvement available to us, I have selected a distinctive set of three – lean thinking, work-out and appreciative inquiry (AI). All are formally defined, have explicit methods, deploy particular tools and can testify to having realised significant organisational benefits.

Lean thinking has benefitted from growing interest and notoriety since it first came to prominence with the publication by Womack, Jones and Roos of The Machine that Changed the World in 1990, and of its sequel Lean Thinking in 1994. In and of itself, the term ‘lean’ is a Western construct. The methods codified as such are generally regarded as having had their origins in the Toyota Production System (TPS). The development, integration and application of lean ideas across industries over the past twenty years has led it to be regarded as an approach to organisational improvement worthy of very serious consideration.

On the other hand, criticisms of lean focus on its ‘tool-head’ status. Indeed, Dan Jones himself has recently commented, “Many of us in the lean community have focused our attention on improving core processes… by deploying brilliant tools when we should have been focused on improving the management process itself.”

But not all lean thinkers have been obsessed with ‘brilliant tools’. Taiichi Ohno in 1978 referred to the Toyota Production System as having been “built on the practice and evolution of the scientific approach”; Steven Spear and H. Kent Bowen in their 2006 HBR article Decoding the DNA of the Toyota Production System describe the effect that the combination of ‘brilliant tools’ and ‘problem solving capabilities’ has in creating “a disciplined yet flexible and creative community of scientists who continuously push Toyota closer to its zero-defects, just-in-time, no waste ideal” and Jeff Liker in his 2004 The Toyota Way has re-enforced the centrality of the scientific method. More recently, Mike Rother, Toyota Kata, has confirmed the trend in understanding TPS as a move away from tools towards systems or routines, of thinking and behaviour.

Toyota’s influence is ubiquitous. Katsuaki Watanabi, in Lessons from Toyota’s Long Drive, describes it thus:

“We’re doing the same thing we always did; we’re consistent. There’s no genius in our company. We just do whatever we believe is right, trying every day to improve every little bit and piece. But when 70 years of very small improvements accumulate they become a revolution”.

Appreciative inquiry couldn’t, on the face of it, be more different from lean. Lean’s insistence on dominant, rational, scientific, process orientated approaches to organisational improvement leaves practitioners of AI cold. Appreciative inquiry is based on the premise that organisations change in the direction in which they inquire. So an organisation which inquires into problems will keep finding problems (and be paralysed by the burden of these) but an organisation which attempts to appreciate the essential nature of itself, at its best, will amplify its ‘positive core’.

Amongst its sources, AI is influenced from personal positive psychology and its origins are generally credited to David Cooperider, who contrasts the notion that ‘organising is a problem to be solved’, with the appreciative proposition that ‘organising is a miracle to be embraced’. A number of sources support the beneficial outcomes that AI might produce. These include Jane Elliot’s work in labelling theory, Open Space Technology and the theories of Social Constructivism. But this evident idealism is also supported by a clear and pragmatic method which is capable of engaging large numbers of people in an enthusiastic and focussed search for organisational improvement and real results.

Detractors of AI have a relatively easy time ‘cocking a snoop’ at it. They point to its heavy investment in activity, to the preference given to social engagement and to the apparent subjectivity of its method: story collecting and telling may well be culturally important and may well give meaning and encouragement to people, but does it really and sustainably make my organisation work any better? AI’s retort, however, is simple – we’ve had decades of ‘rational-scientific’ approaches to change and most of it hasn’t worked.

As Whitney and Trosten-Bloom point out in The Power of Appreciative Inquiry, AI works because it liberates power. It unleashes both individual and organisational power, and generates unprecedented cooperation and innovation. It is a positive revolution that can’t be stopped.

Work-out is ballsy and explicit. Unlike lean thinking, it focuses little on detailed and sophisticated diagnostics; unlike AI, it focuses less on ‘what I need to become’ and rather more on ‘what performance I need, now’. Work-out has its origins in the problem of ‘busting bureaucracy and engaging the front line’ in GE under Jack Welch. Its notoriety has been somewhat eclipsed in recent years by GE’s reputation for six sigma. But work-out is generally credited with bringing about much needed transformation towards speed, simplicity and self-confidence inside GE during the earlier years of Jack Welch’s stewardship and, as an improvement method, beyond its Crotonville origins, is capable of citing many examples of performance and cultural benefits.

Where lean has a reputation for focussing on the process, and AI a reputation for engaging the community, work-out is concerned first and foremost with results and with achieving these in very short timescales. Its own critique of other change methods is that they are slow, unfocussed and activity-centred. Who cares how many people might have been trained in quality improvement methods, it argues, if there is no sensible return on investment?

When organisational leaders invest time in defining the challenging improvements required in their operations, it becomes possible to work backwards from desired results and to make accountable, most usually, front-line people in working out how such results can be realised. This focus on results is an enduring one, as is the need to deliver them speedily, in usually not more than ninety days. By working in this way and by requiring managers to make rapid on-the-spot decisions regarding the implementation plans of their people, managers are compelled to delegate appropriately and to trust their people to deliver. This serves to create a dynamic leadership capability and to eliminate, as was its original intention, slow, bureaucratic and analytical behaviour.

Critics of work-out point to its intensity, arguing that galvanising the necessary stakeholder energy is sustainable only over short periods. In this sense, they argue, work-out merely represents a ‘shot in the arm’.

But work-out, say its advocates, is about developing leaders who have new expectations about the pace at which things can be done – busting bureaucracy means ‘taking it on’, not ‘having tea and biscuits’ with it. Work-out’s revolutionary garb merely re-asserts the natural, prebureaucratic order of things, moving improvement behaviour from being, in Steve Kerr’s words, “an unnatural act in unnatural places to natural acts in natural places”.



So, what conclusions might we draw from having compared these three distinct approaches to organisational improvement?

Firstly, each approach arose from real and deep seated problems – for Toyota it was the desire to compete with Ford on a platform other than unaffordable mass production; for GE, it was the need to reduce the bureaucracy of organisational practices; for the AI movement it was the dissatisfaction felt at the poor success rate of deficit-based change. So, perhaps, for improvement methods to stick, their need must be viscerally relevant; their design must have integrity and be anchored in the needs and culture of the organisation.

Secondly, each approach required, a priori, that the natural capacity of people to manage risk, solve problems and realise opportunities, be leveraged. For Toyota, finding ways to make good cars with, in comparison to Ford, only a tiny proportion of the necessary resources, required the ingenuity of all of its engineering talents; for hierarchical GE, creating a leadership system which would bring front-line ideas to fruition and ‘bust bureaucracy’ required the deep involvement of people at all levels; for the AI movement, the failure of conventional change with its narrowly applied expertise and focus on ‘what doesn’t work’ required the perspective of the wider organisational system (all of the people) in identifying and building from the positive core of practices. So, perhaps for improvement methods to stick, there must be a deep and necessary requirement that people’s ingenuity be marshalled.

Thirdly, and perhaps most significantly, leveraging this natural capacity was made possible by surfacing and then resolving the many issues that sit inside the original (deep seated) problem. For Toyota, this is done through the design of processes that make front-line problems visible in real time; for GE, it is through the framing of challenging opportunities by leaders for improvement by their people; for AI, it is through the definition of strategic themes around which communities of people build on ‘the positive core’.

Perhaps, without this disposition to make the issues explicit, to surface them, there will be little progress; resolving is a matter of technique; surfacing, by contrast, is a matter of personal psychology. Surfacing requires leaders (and their organisations) to want to find issues, to have a liking for so doing and to be able to cope with the uncertainty that such uncovering engenders. It requires a shift in the emphasis of leadership discourse and practice away from a focus on the persuasiveness of advocacy and the control of admonishment towards the illuminating effect of inquiry. Without this, no amount of tools based improvement will ever sustain.


Additional insights and questions may be yielded by further understanding of and comparisons between other improvement methods (six sigma, positive deviants, systems thinking, kaizen events and so on). Comparative analysis (at a level deeper than this article has been able to go) has a solid reputation as a research method and for practitioners is a viable way of thinking differently and thinking again about what might really work inside their organisations.

And so, our ‘call to arms’ is this: explore the diversity of thinking and experience on organisational improvement that exists; consider combinations of approaches which meet the deep issues that you need to address and run experiments which help you to understand what may or may not work. By all means, embrace the transformational potential of lean but challenge narrow and repetitive codification. Lean didn’t start as a package; why should it become one?

In the end, Mr Lenin, perhaps we get the revolution we deserve.


Kevin presents three approaches to improvement, claiming they are distinctive. No doubt, two of them (lean and AI) are widely known and practiced, with large numbers of references, but the third (work out) is much less known, with only one specific 2002 book listed on Amazon.

But why should lean and AI be distinctive? Certainly they have tended to be used in different fields – lean starting out in manufacturing and growing into service, with AI more or less the other way around, starting particularly with service. AI now seems to be merging with lean. AI is strongly associated with organisation development, which certainly has a role in every lean transformation.

Both lean and appreciative inquiry share at least two characteristics – systems thinking, and management by asking questions. Moreover, are ‘people’ centred and future vision centred. But, certainly, lean is more focused on ‘logical’, closed problems whereas AI’s field is more open, where ‘the problem is the problem’. Of course, both types are routinely found – that is why an awareness of both is useful. For readers of this journal, who will be familiar with lean but perhaps not as familiar with AI, I would recommend ‘The Thin Book of AI’ as a publication that will help in your change endeavors (it is recommended on the MSc in Lean Operations, when AI is discussed).