Rob Powell

Link to article –> here

• Product development is not analogous to manufacturing - tasks are (mostly) highly variable and not predictable. Adopting a factory approach to product development is counter-productive.

• Product quality decreases when resource utilisation is pushed close to 100% by management. Managers don’t appreciate the negative impact of high utilisation.

• High utilisation leads to queues which affect economic performance. Due to the nature of the work, it is difficult to observe queues and product work inventory - there are no physical signs.

• Better to provide a capacity buffer in processes that are highly variable - e.g. Google 20% days. Easy to say, not so easy to implement! Better to target teams and managers on productive outcome rather than high resource utilisation.

• A reduction in active work can help to sharpen focus, as can making WIP visible. Reducing batch size cuts WIP and speeds up feedback which improves cycle time, quality and efficiency.

• Batch size should strike a good balance between transaction costs and holding costs. In software/product development, this is the difference between large batches of code every few months to testing smaller batches a few times a day.

• Product work is highly variable and so trying to obstinately follow a plan is not appropriate. Plans should be treated as hypotheses that are subject to continuous evolution as we continuously learn.

• Less can be more. ‘Simplicity is the ultimate sophistication’ - Leonardo da Vinci.

• Problem definition is the hardest and most important part of the innovation process. Invest time here as it sets good context for everything that follows.

• Users usually want a solution that ‘just works’.

• The push to ‘get it right first time’ should be replaced with a tolerance of ‘getting it wrong first time’. Getting it wrong early is cheaper than finding a problem much later on when the cake is almost baked.

• Product development is not analogous to manufacturing!