Library · book

Lean Software Development: An Agile Toolkit

Mary Poppendieck & Tom Poppendieck
2003·Addison-Wesley

Source: https://www.informit.com/store/lean-software-development-an-agile-toolkit-9780321150783

The Poppendiecks translate Toyota's manufacturing principles into software terms: eliminate waste, amplify learning, decide as late as possible, deliver as fast as possible, respect people, see the whole.

The translation is literal enough to be rigorous and loose enough to be useful.

The seven wastes of software (partially done work, extra features, relearning, handoffs, task switching, delays, defects) are the single most portable diagnostic tool a product director can carry into any team.

Read alongside Reinertsen for the economic argument and Anderson for the implementation. It is not motivational literature; it is a manual.

Central argument

The Poppendiecks argue that Toyota's lean manufacturing principles translate directly into software development practices, and that the primary lever for improving software teams is the systematic elimination of waste — defined across seven specific categories including partially done work, extra features, handoffs, and task switching. Their central thesis is that software development is not a production process but a learning process, and that traditional project management disciplines misapply factory-floor controls to work that is fundamentally about acquiring knowledge. The toolkit format is deliberate: each principle is paired with concrete practices and decision tools intended to be applied by teams, not just studied.

Critique

The manufacturing-to-software translation, while intellectually rigorous, quietly inherits an assumption that may not hold: that the unit of value being optimised is throughput of working software. This can create a tension when applied to product discovery, where the goal is often to slow down and generate learning before committing to build — a phase where 'delivering as fast as possible' is precisely the wrong instinct. The book was written in 2003 against a backdrop of over-engineered waterfall processes, and some of its arguments lose nuance when applied to modern product organisations where the bottleneck is not delivery speed but validated direction.

Why it matters for product

The seven wastes framework is directly actionable for a CPO diagnosing why a product organisation is slow or misaligned: partially done work surfaces in oversized backlogs and unshipped features, extra features reveal a discovery process that doesn't kill ideas early enough, and handoffs expose team topology problems that no amount of agile ceremony will fix. The principle of deciding as late as possible has direct application to roadmap governance — it reframes premature commitment not as indecision but as economically rational preservation of optionality. Read against Reinertsen's cost-of-delay logic, it gives product directors a vocabulary to argue against speculative roadmaps at the executive level.