I was going to write a more traditional book review for The Lean Startup (2011) by Eric Ries, which I just finished listening to today, until I got to the epilogue. Let me give you a bit of background, and then I'll explain why I'd like to open this up to a broader discussion.

The Lean Startup promotes ideas which Ries has been developing since at least his time at IMVU around how startups should be nurtured and managed. Ries defines a startup as

"...a human institution designed to deliver a new product or service under conditions of extreme uncertainty."

And, throughout the book, applies this label to a variety of organisations, including stereotypical tech startups, but also government agencies and manufacturing. An institution doesn't need to be creating the next hot social network or IoT gadget to fall under this definition. Any group of people that are trying to break new ground and deliver something new can learn from this book's advice.

Ries argues that startups should seek to achieve validated learning, wherein changes to a product or service are dictated by indirect customer feedback, and not by hunches or predefined plans. As Ries stresses throughout this work:

"...planning and forecasting are only accurate when based on a long, stable operating history and a relatively static environment. Startups have neither."

"Indirect" because consumers often don't really know what they want, or don't know how to express it properly. "Validated" because the approach should be scientific -- develop a hypothesis, carry out experiments, and determine whether your hypothesis was correct or incorrect

"If the plan is to see what happens, a team is guaranteed to succeed -- at seeing what happens – but won’t necessarily gain validated learning. If you cannot fail, you cannot learn."

A/B testing, MVPs, and the lean methodology were revolutionary only a few decades ago, but are now the default mode of working for organisations looking to break new ground. In this sense, The Lean Startup can be seen as a sort of spiritual successor to The Mythical Man-Month, whose ideas were also unorthodox at the time, but are now mainstream.

But the end of the book is what really got to me.

In the epilogue, Ries talks about the precursor to the lean methodology movement -- scientific management. In the late 18th and early 19th centuries, Frederick Winslow Taylor, an American mechanical engineer, sought to improve the efficiency of businesses. Dividing businesses up into functional departments, dividing workloads into tasks, streamlining production lines to reduce waste and inefficiency -- all of these are obvious today, but less so in Taylor's time. "Taylor effectively invented what we now consider just 'management'", notes Ries, as well as "the idea that work can be studied and improved through conscious effort."

"The revolution that he unleashed has been -- in many ways -- too successful. Whereas Taylor preached science as a way of thinking, many people confused his message with the rigid techniques he advocated. ... Many of these ideas proved extremely harmful, and required the efforts of later theorists and managers to undo."

Ries goes on to recount an anecdote told to him by someone who had attended one of his recent conference talks. This person took Ries' advice to heart and promoted validated learning, the five whys, and other aspects of the lean startup within his business. As a result, he gained a reputation as a brilliant engineer within his company. But his superiors didn't actually learn to follow lean methodology, despite his proselytising -- they simply thought they needed to improve their hiring process to find more "10X engineers", like him.

I fear that this story is a microcosm of lean methodology as a whole, as it is interpreted and applied today.

Just as those in Frederick Taylor's time couldn't see the forest for the trees -- rigidly applying the techniques of scientific management without fully understanding or appreciating their motivation, or significance -- some startups today apply "lean methodology" without really understanding what that means.

I sometimes get the impression that Gantt charts, Kanban boards, and Jira tickets -- while still very useful when used correctly -- have simply become a way of signaling "we do lean development", without actually following lean principles.

At their cores, both scientific management and the lean methodology are driven by a scientific approach to understanding and improving the development of products and services. But I get the feeling that, while some companies are doing actual science, others are just filling their laboratories with equipment without ever performing any experiments.