The Relevance of Giants – 1. Deming

The Relevance of Giants – 1. Deming

On most every occasion when I’m speaking in public – at conferences, workshops, and the like – I tend to mention one or more of my “Giants” of Rightshifting. Men and women who, through their lives and work have contributed significantly to my understanding of work, and in particular to my understanding of effective collaborative knowledge work.

Many folks express interest in these Giants, but I do wonder if they appreciate the relevance of the ideas and experiences of these Giants to their own daily lives at work.

I mean, what relevance does, say, Bill Deming have to developers, testers, operations staff and the like? Which aspects of any of these Giants’ work could be useful or helpful or simply comforting to these folks?

In this occasional series of posts I’ll be exploring some of the Giants’ relevance to folks other than theorists, managers, consultants and the like. I’ll be sharing some insights into their work, and specifically, the likely relevance.

With these posts I hope to pique your curiosity just a little. Let’s start with Bill Deming.

W. Edwards Deming

Bill Deming

(October 14, 1900 – December 20, 1993)  (See also: Wikipedia entry)

I’m not going to dwell on his work in SPC (Statistical Process Control) or SQC (Statistical Quality Control), his pivotal role in the Japanese post-war economic miracle, his 14 Point system of thought he called the “System of Profound Knowledge”, nor his Plan-Do-Study-Act (PDSA) cycle (the latter being the basis for most Agile approaches, btw).

Deming’s 95/5

I suggest the primary relevance of Deming to most folks working in the field of software development (and production operations) is primarily the idea known as “Deming’s 95/5” (although this originated in a quote from Peter Scholtes).

“The fact is that the system that people work in and the interaction with people may account for 90 or 95 percent of performance.”

From my studies of Deming, and from applying his ideas in my practice, I have come to believe that it’s the interactions between people that account for the lions share of “productivity”, “performance” and “success” in collaborative knowledge work. And the “system” a.k.a. the way the works works has a major (hidden) influence on the quality of those relationships, as well as on the work (output, results) of the individual workers.

“Dr. Deming taught me that 95% of the performance of an organization is attributable to the system (processes, technology, work design, regulations, etc.) and [only] 5% is attributable to the individual.”

~ Tripp Babbitt

Where’s the Relevance?

If, like most people, you’re looking for a better quality of life at work, Deming points the way to us improving our relationships with our colleagues, peers and managers. Maybe this perspective is something to consider on those occasions when you’re less than happy in your work, when you’re checked-out, or disengaged, or frustrated.

And Deming’s attribution of 90-95% of your performance to the system within which you’re obliged to work throws a new light on many typical organisational practices such as history-led recruitment, performance appraisals and reviews, stack ranking, criticisms (and praise) for your efforts, etc.. Your results (and self-esteem) may be taking a hit from the effects and constraints inherent in that system, not from anything you’re doing (or not doing) yourself.

Practical Investigation

Deming designed the Red Bead Experiment to illustrate these very points, in a way that most people can directly relate to.

– Bob

Further Reading

Four Days with Dr Deming ~ Latzko and Saunders
95% of performance is governed by the system ~ Vanguard web page

5 comments
  1. antlerboy - Benjamin P Taylor said:

    Bob, I don’t know how helpful this is, but for some reason, I’ve always been tempted to question the cult of Deming. I don’t want to detract from the man or from his wisdom, or the enormous value of practicing what he taught. It’s just that the uncritical lines which are the ‘received wisdom’ about his life and approaches do, I admit, bother me a bit. I sure wish I had been to his seminars, but I feel that digging deeper her might be more interesting than the headlines. This has turned into a much longer comment than your post – but perhaps is interesting, particularly on your theme of ‘the relevance of giants’!

    So:
    – “his pivotal role in the Japanese post-war miracle”. Yes, undoubtedly – we have the evidence of the Japanese recognition of this. And yet, your wiki link intriguingly /doesn’t/ mention Deming – and credits two main reasons: successful economic reform by the government and the outbreak of the Korean War… the former point, however, amongst a lot of intriguing stuff, includes Keiretsu, ‘mirroring wartime conglomerates, or zaibatsu’, which spurred both horizontal and vertical integration – precisely the sort of message Deming preached, as I understand it – but a result, or a happy coincidence? The rest, including ownership and loan measures, would reward richer study.

    – his Plan-Do-Study-Act cycle. Yes. But I think he always talked about this as the Shewhart cycle (I have a lovely photo of his original overhead transparency, including this phrase, from one of his interns – who, incidentally, also said – if I heard right – that she slept in his hotel room, at the foot of his bed, on his many teaching tours). I think it’s interesting and valuable to situate him in the context of others he learned from…
    Your comment about the link to agile is also interesting – intuitively, it makes great sense, of course – however, is there an explicit, intentional link that you’re aware of?

    What’s perhaps interesting as well is the ‘organisation as a system’ diagram (Deming acknowledged that he didn’t know ‘systems thinking’, and was talking about processes – a confusion which has left a bit of a legacy – but I think ‘processes as a system’ are interesting) – https://blog.deming.org/2012/10/appreciation-for-a-system/ – and Deming’s insistence, in his Japanese lecture (translation http://hclectures.blogspot.com/1970/08/demings-1950-lecture-to-japanese.html), that PDSA must be applied not only to the the traditional three steps:

    1. Product design (shoes, cotton materials, silk materials, magnetic products, electrical appliances)
    2. Manufacture
    3. Sales

    (It’s interesting he chooses this instead of Shewart’s “three-step process of specification, production, and inspection” https://en.wikipedia.org/wiki/PDCA#About )
    But…
    “Today, in 1950, we all must design, manufacture, and sell in the same way, but science has expanded. Rather than following the example of these three steps, four stages, including market surveys which I have just highlighted have become necessary.”
    I think he designed and sold the ‘market surveys’ – that’s interesting…

    – “Deming’s 95/5” – I have it on good authority that the original statistic was based on a study of work in a US prison, published in 1911 from research done much earlier. And the wiki you link to – https://en.wikipedia.org/wiki/W._Edwards_Deming – including the intriguing line “He told Ford that management actions were responsible for 85% of all problems in developing better cars.”
    And, anyway, the 95% is partly about the people anyway… so this is intriguing, really, worthy of further exploration…
    (The Tripp Babbit quote is nice – I see he picked up the John Seddon phrasing; it seems to imply that they were taught directly, which I am not sure is the case!)
    Two downsides of the ’95/5 rule’ – as I’ve seen it cast – are (1) a serious risk of denial of individual responsibility and (2) a tendency to ‘blame’ (a fairly non-systemic concept, I’d say) ‘management’ instead of ‘workers’. Neither very helpful, really…

    This looks like a good sourcing of 94/6 from ‘Out of the Crisis’ and 90-95 from Deming’s introduction to Scholtes’ Team Handbook (not Scholtes himself) – https://management.curiouscatblog.net/2013/04/24/94-belongs-to-the-system/

    “Red bead experiment” – it’s cool, I’ve participated and facilitated. It makes some very good points. It could be argued that it’s a ‘lie to children’ – a simplification for the purposes of teaching, for rhetorical effect if experiential learning can be considered rhetorical. But it therefore, by exaggerating the conditions, threatens to weaken the real deep long-term learning. According to https://www.spcforexcel.com/knowledge/variation/red-bead-experiment, “Dr. W. Edwards Deming often referred to it as a stupid experiment that you’ll never forget.”
    BTW, how and who developed the red bead experiment, I can’t quite be sure. This – http://www.redbead.com/what/ – says “In 1982, a teaching tool was created with Dr. Deming that he used in his seminars around the world to teach his famous 14 Obligations of Management.” – ‘created with’ sounds really interesting!

    Thanks for indulging my exploration here – two more points to make from googling and trawling these links:

    1) Christianity – I never knew that he and Scholtes were big Christians, and evidently both wrote devotional music. That’s intriguing!

    2) There’s an even bigger issue – what is the *real* history of post-war quality in Japan? Deming was certainly recognised and honoured. But Myron Tribus and Ken Hopper, serious historians and documenters of the quality revolution in Japan and later in the US, report that Sarasohn probably deserves more credit (along perhaps with Protzman… and Shewhart)

    “Homer Sarasohn… the mad who made Japan successful”
    http://www.firstmetre.co.uk/library/documents/493/

    “Had Deming and I stayed at home, the Japanese would have achieved world quality leadership all the same.”
    (Joseph Juran) in https://link.springer.com/chapter/10.1057%2F9780230282179_15

    Superb article: Quality management and quality practice: Perspectives on their history and their future
    https://deepblue.lib.umich.edu/bitstream/handle/2027.42/61868/756_ftp.pdf?sequence=1

    “Why haven’t we heard much of poor old Homer? Because of the American quality guru W.Edwards Deming, a master of self-promotion, who was brought to Japan to continue the work of Sarasohn and Protzman. While Sarasohn was MacArthur’s man, Deming was perceived as the Japan Union of Scientists and Engineer’s man, which helped ensure his place in history.”

    The comments on this economist ‘guru’ article on Deming (https://www.economist.com/node/13805735/comments) are enlightening!

    Kenneth Hopper himself says:
    “In four years this democrat in dictator’s clothing (Homer Sarasohn) may have accomplished more than any economic dictator in history” “A lesson learned and a lesson forgotten”, Robert Wood, FORBES.
    “Sarasohn teamed up with Protzman in 1948 to design and teach intensive management training seminars, the Civil Communication Section (CCS) Management Seminar .. After World War II, Japan’s ‘captains of industry’ fortified Sarasohn’s management values with Japanese Bushido values.. , they … produced the postwar ‘miracle’? Yoshio Kondo, Baruch College, City of New York.
    “A leader’s main obligation is to secure the faith and respect of those under him”, Charles Protzman and Homer Sarasohn, CCS Manual, GHQ Tokyo, 1950
    “I gave over 500 lectures in every part of Japan from Hokkaido in the north to Kyushu in the south” “CCS was the light that illuminated everything” “By the end of the (1950) CCS seminar we all knew we would catch up with the Americans”
    Bunzaemon Inoue, co-chair, CCS Seminars. Chairman, Sumitomo Rubber Industries
    General MacArthur’s Civil Communications Section was very important for both Deming and Japan’s success. Our ignorance of it is surprising. CCS drilled into Japanese top executives that they were responsible for good management. When one missed a CCS session, CCS was on the phone to remonstrate. “Our slightest wish was their command”, Protzman remembered(1) Deming would be saddened at CCS neglect. I know. When illness struck, he sent me a substantial sum to help me tell the CCS story. When Human Resource Management published my “Creating Japan’s New Industrial Management” Deming wrote me “This is just what I need.”
    Late in 1969 Peter Drucker phoned to say he had someone I should meet. I asked no questions and turned up to find a near mythical Japanese Sensei (teacher) Takeo Kato. Kato started to tell me about CCS when the tall Polkinghorn appeared and Kato said, “No man has done more for Japanese industry”. I was hooked. In 1979, my wife and I were treated like royalty in Japan at the invitation of Sumitomo Electric, Sumitomo Rubber and Matsushita Electric. Because we knew the great CCS engineers, dinners, waiting limousines, guides and interpreters were everywhere. The story of CCS is now well known in industry but has not reached business schools who prefer to see Japan’s success as an inexplicable Miracle. Simplistic Shock Treatment was given Russia when the lesson from Japan was rebuilding requires work in depth including improving how factories run. The most let down were non Asian developing nations.
    Communications in Japan in 1945 were a disaster. The War Department approached US industry for help and able people responded including a young Homer Sarasohn who had impressed the US Army at MIT’s Rad Lab, Charles Protzman, a 6.ft 4in manufacturing superintendent with decades of experience from Western Electric and Frank Polkinghorn a high engineer in Bell Labs. I came to know all well. Their ability and domain knowledge made it possible for Japan’s unimpressive electrical manufacturers to become its world stunning Consumer Electronics Industry. Influenced by Morgenthau, the US had shackled MacArthur with Secret Order JCS1380/15 to take no responsibility for the Japanese economy. As a result, when Truman announced the Reverse Course in late 1948, CCS was the only Section engaged with manufacturing. CCS proposed that it combine seminars with working with its manufacturers to help them compete in world markets. Sarasohn loved to recount his 1949 confrontation before MacArthur with the large Economic and Scientific Section who argued the US would be giving away too much. MacArthur sat expressionless through both presentations, got up and walked to the door. Sarasohn thought, “I’ve blown it”. MacArthur turned, pointed the stem of his corncob pipe at him and said, “Go do it”. The rest is history. Japan’s electronics industry would have a major influence on management in the rest of Japanese industry and the world.
    Japan’s specialists wanted a visit from Shewhart. Sarasohn refused until 1950. When he phoned, Shewhart declined for ill health. Sarasohn confirmed Deming’s ability. At his request, ESS issued the formal invitation. Many have confirmed CCS importance. My brother Will has a selection at http://www.puritangift.com including an abbreviated chapter on CCS and Japan’s extraordinary executives from our book The Puritan Gift, a Financial Times Top Ten Business Book of 2007 Ken Hopper
    “A lesson learned and a lesson forgotten”, Robert Wood, FORBES, Feb 6, 1989. “Homer Sarasohn and American Involvement in the Evolution of Quality Management in Japan, 1945–1950”. N. I. Fisher, International Statistical Review (2008). K. Kobayashi, Chairman NEC Corporation, in his 1985 address to Bell Laboratories. “For more about the early development of quality control in postwar Japan, see ‘Quality, Japan and the U.S.: the First Chapter’” Ken Hopper Quality Progress Sept 1985

    And, even more intriguing to me, it could be that Elliott Jaques’ Requisite Organization theory might have played a valuable role:
    Ken CraddockJun 13th 2009, 19:25
    The Japanese also adopted Elliott Jaques’ requisite theory to counter upper-level organizational design, long-term appraisals, and executive selection and development. Deming consulted and taught four days per month for 10 years at GM with little apparent effect. The two theories are robust to each other – but only if both are implemented. Treating either as a short-term fad negates it. Does any US-based Board of Directors know what to do and how to do it?

    Excellent downloadable article:
    https://www.researchgate.net/publication/46537675_Homer_Sarasohn_and_American_Involvement_in_the_Evolution_of_Quality_Management_in_Japan_1945-1950

    Whole site trying to redress the balance: https://honoringhomer.net/

    Brilliant overview by Bob Cringely: http://qualiticien.over-blog.com/article-679016.html

    More or less the same piece by someone else: http://www.richmore.com/html/protzman.html

    Another overview: http://quality-history.blogspot.com/2007/11/how-general-macarthur-allowed-japan-to.html

    https://en.wikipedia.org/wiki/Homer_Sarasohn

    Myron Tribus video interview:
    https://blog.deming.org/2017/02/myron-tribus-interview-of-homer-sarasohn/

    I hope this illustrates both that the footnotes might be as interesting as the headlines – and that there might be more to the Deming story – and more interesting things there that are *not* Deming.

    • Thanks for this. What would you say is Deming’s relevance to eg developers, testers and the like?

      • antlerboy - Benjamin P Taylor said:

        Well, that’s an interesting question. I’d say straight up front that I know very little about those folks and their worlds! I’m more in service transformation and other organisational issues. So this is just farting around 🙂

        But some things suggest themselves, using a list from your outline and the comments I made, rather than Demin’s 14 points etc (though that would be an interesting exercise, I think):

        – Integrated supply changes and collaboration – what would vertical and horizontal integration mean in the software world? I would suggest it would go all the way back to those who develop programming languages, and hardware – and certainly (DevOps, I suppose) be as much about the organisation of the business – especially any helpdesk/customer services type function – as the software

        – Proper PDSA with multiple opportunities for testing in the real world – I recognise this is pretty core for agile so I won’t labour it! But what would applying, say, an A3 approach to development and testing look like? I recognise the various pre-emptive roles of testing used in some programming approaches, but PDSA (I believe) requires hypothesis development and testing – is this a thing? Could/should it be? I suspect it is implicit, but making it more explicit could be interesting.

        – Take a leaf from the lecture in Japan and consider what the ‘whole system’ for programming/software would be – how do you engage with the market and analyse need? i.e. the whole lifecycle from market to every stage of development to results, and back round again. I recognise that lean startup is quite well engaged on this and that software tends to quite naturally align the ‘build and sell’ cycle with the ‘research and develop’ cycle (well, apart from the large-scale software providers and most peoples’ experience of software in practice!)

        (as a minor diversion, it would be interesting to explore some of those ‘reasons for the Japanese miracle, and see how they might apply – especially the financing elements. I’ve always thought that one of the key factors behind the development of ‘lean’ was capital scarcity – not completely sure now this is the case – but certainly the need to reduce, reuse, and repurpose heavy capital (e.g. there are still stories of Toyota factories using beautifully maintained machinery from the 60s and 70s alongside modern robots). What would this mean in software? Certainly modularity – when I was a young nerd and worked in software, the modularity and development of sub-routines etc at recursive levels was one of the things that really excited me. And I recently saw a good tweet – maybe on your timeline – recommending talking about ‘infrastructure investment’ rather than ‘technical debt’)

        (and another diversion, from the two above points – to what extent is lifecycle of software actively considered and developed from the start?)

        – Work on the system as well as the people – retrospectives attempt to cover this, but is there really a learning loop about the processes followed? Perhaps for the core programming part, if you’re lucky, but for the ‘whole system’?

        – Understand variation – is there any real understanding of statistical analysis of user research and interaction research/design? Especially for live systems / websites etc?

        Don’t know if any use, but a fun thinking exercise 🙂

Leave a comment