HomeVoices › W. Edwards Deming
W. Edwards Deming
Worth following

W. Edwards Deming

Quality, systems and why blaming people fails

Taught post-war Japan that quality is a property of the system, not the workforce — and was ignored at home for thirty years.

USA

Photograph: FDA · Public domain · Wikimedia Commons

Deming was a statistician who argued that most variation in quality comes from the system a worker operates in, not from the worker. His work was taken up in Japan in the 1950s, where it contributed to the transformation of Japanese manufacturing's reputation, long before American industry paid attention.

The PDSA cycle — Plan, Do, Study, Act — is his, as is the argument that inspection at the end is the wrong place to build quality. His 14 Points for Management and the Seven Deadly Diseases are direct, uncomfortable and still land.

The bit that matters most

'Drive out fear.' He argued that people who fear the consequences of reporting a problem will hide it, so the organisation loses the information it most needs. Every quality system that relies on people volunteering bad news depends on this and most ignore it.

What they teach

The system produces the results, not the people

If you want different results, change the system. Exhorting people to try harder inside a system that produces errors is, in his framing, both futile and unfair — and it is the default management response.

Understand variation before you react

Some variation is normal noise; some signals a real change. Reacting to noise as though it were a signal — the good month, the bad week — makes performance worse, not better. This is the statistical heart of his work.

Stop awarding business on price alone

One of the 14 Points. Buying on lowest price ignores the total cost of poor quality, late delivery and rework, and it destroys the supplier relationships that would fix them.

Where to start

  1. The W. Edwards Deming Institute — The 14 Points, the PDSA cycle and the primary material, free.
  2. Deming's 14 Points — Read them straight through — they are blunter than the management literature they inspired.

These links go to their own work. We summarise and point — we do not republish it.

Who it suits

Anyone running an operation, and any manager whose instinct when something goes wrong is to find out who did it.

And who it does not

The statistical process control material is technical, and the tone is prophetic. It rewards patience.

More voices