Out of the Crisis

by W. Edwards Deming · 9 ideas

  1. Quality Lowers Cost

    Many people assume better quality costs more. In fact, work done right the first time needs less rework, less waste and fewer delays, so costs fall and output rises. Improve the process, and you raise quality and save money at once.

  2. Two Kinds of Variation

    Results vary for two reasons: ordinary ups and downs built into the system, and rare events with a specific cause. Each needs its own response. Look for a cause only when a result falls far outside the usual range, and improve the system to shrink the rest.

  3. Don't Chase Every Wobble

    Adjusting a process after every result that misses the target feels responsible. But reacting to ordinary ups and downs stacks your corrections on top of them, so results swing even more. Leave a stable process alone unless you are changing the system itself.

  4. Build Quality In

    Checking finished work for defects means you pay once to make the mistakes and again to find them. Inspection sorts good from bad, but it never makes the next batch better. Fix the process that produces the defects, so there is less to catch.

  5. Slogans Change Nothing

    Posters and pep talks that urge people to try harder assume the problem is effort. Most mistakes come from the tools, materials and methods people are given, which they cannot change themselves. Skip the slogans and fix what only managers can fix.

  6. Targets Without Methods

    A number to hit, with no change to how the work gets done, leaves people only bad ways to hit it. Afraid of missing, they cut corners, hide problems or bend the figures. Before you demand a better result, improve the process that produces it.

  7. Rankings Measure the System

    When people do the same work in the same system, most of the gap between their results comes from the system and from chance. Ranking them punishes bad luck and teaches people to compete instead of help. Work on the system, not the scoreboard.

  8. Total Cost, Not Price

    The cheapest supplier often costs the most once you count defects, delays and rework. Switching to whoever bids lowest also throws away the trust and shared learning built with a good partner. Choose suppliers on total cost and stay with the good ones.

  9. Invisible Figures

    Some of the most important numbers can never be known, like the business lost through one unhappy customer or gained through a happy one. Managing only by the visible figures treats these as zero. Give real weight to effects you cannot measure.

More on systems