They pressed on, furiously forcing simplification and standardization. An inventory of more than 1,200 sizes of metal tubing was reduced, in one fell swoop, to 50 standard sizes. Everyone howled, but it worked. Wrench sizes were standardized. These live on in hardware stores today, marked “SAE” for Society of Automotive Engineers. Within months, thousands of individualistic machinist-craftsmen were forced to simplify and standardize their work. Automation followed this industrialization. Assembly times for autos fell by an incomprehensible 70 percent.
By 1917, advocates of industrialization methods for office work emerged. W.H. Leffingwell successfully industrialized a large advertising organization, achieving a fourfold increase in total productivity. Improvements included a simplified office workflow and standardized forms, ads and job activities. These were disruptive changes — forced upon the workforce. Office machinery was brought in afterward. But managers rejected this successful, disruptive industrialization approach.
The knowledge work industrialization revolution
Office automation producers have always struggled to prove the business case for their products. In the early 1990s, scholars began to refer to the economy-wide failure of computer technology to increase productivity growth as “The Productivity Paradox.”
But the true paradox is the fact that management continues to try to “subtly force” standardization of knowledge work with technology, even after a century of failure. The success of disruptive industrialization techniques is rejected. And now a new wave of technologies is again expected to “subtly force” standardization without disrupting the knowledge work status quo.
Does any of the following look familiar?
Automated error correction: Does your company maintain vast pools of inconsistent data that require tedious manual reconciliation and are therefore prohibitively costly to analyze? Then try out a new “data wrangling” technology platform from any of several Silicon Valley startups. These promise to robotically integrate, analyze and visualize those pesky inconsistent data elements.
As a disruptive alternative, you could invest a fraction of that cost to create and manage interchangeable data elements, or “parts,” for your knowledge work assembly lines. The majority of data wrangling involves mundane tasks, such as normalizing inconsistent data fields. Customer names, for instance, are sequenced differently: Some are first name, last name; others are last name, first name. Like manufacturing industrialization, this requires early, disciplined planning and documentation as the data elements are born. But even if it gets out of control, like the proliferation of tubing sizes in Ford plants, it can quickly be forced into a manageable number of standards.
Institutional memory: Worried about knowledge workers joining and leaving your company so frequently that institutional memory is hard to retain? The classic disruptive approach would be to standardize, document and catalog the informal “tribal knowledge” that they use every day. But if you prefer to “subtly force” standardization with a high-tech workaround, you could purchase and deploy a new “knowledge multiplier.” It transcribes conference calls and compiles content. The system’s artificial intelligence then identifies similar challenges that workers face and facilitates connections between them. What could possibly go wrong?
Sign up for CIO Asia eNewsletters.