Some work will not hold still long enough to be drawn as a flowchart. A claims adjuster, a physician, an investigator, an architect — each begins a case without knowing what the next step will be, because the next step depends on what the last one revealed. Traditional business process management assumes the opposite: that the path can be modelled in advance and the worker routed along it.
Keith Swenson and his contributors argue that this assumption is why decades of workflow technology failed the people it was meant to help, and they set out what replaces it. Adaptive Case Management moves the design of the process into its execution. The knowledge worker assembles the path toward the goal rather than following one laid down by an analyst months earlier.
The book is a collaboration among industry experts who ordinarily compete — Nathaniel Palmer, Jacob Ukelson, Tom Shepherd, John Matthias, and others — which makes it unusually candid about where the existing tools fall short. It explains why previous technological approaches did not meet the need, identifies the gap, and lays out the options for closing it.
Written in 2010 for managers of human knowledge workers, it has aged into something its authors did not plan. The problem it describes — coordinating long-running work whose shape emerges as it proceeds, across participants who each hold different pieces of the picture — is precisely the problem of putting AI agents and people on the same task. Anyone building agentic workflows today is rediscovering, often painfully, what this book set out fifteen years ago.
Read it with its sequel, Taming the Unpredictable, which follows with award-winning case studies and practical guidance. This volume tells you why the old approach fails; that one tells you what people built instead, and how it went.
