# Faster Cheaper Better **Michael Hammer and Lisa Hershman**  --- _Most organisations don't know what they're actually doing, and the fix isn't trying harder._ When you map an end-to-end process, you find ninety-four steps. You can then classify each one: value-adding (what the customer will pay for), non-value-adding (checking, tracking, scheduling nobody asked for), or waste (duplicate efforts, errors, reports no one reads). In the company Hammer studied, fewer than 15% of steps were genuinely value-adding. The rest was overhead or pure waste. That ratio isn't a motivation problem. It's a design problem. The way we operate today is a legacy of the Industrial Revolution: work divided and subdivided into smaller and smaller units across functional departments, each optimising locally whilst destroying end-to-end performance. The shipping manager whose job was so narrowly defined that doing it well hurt the company rather than helped it. That's what the [[Execution trap]] looks like in structural form: what appears to be a people failure is almost always a design failure. Nobody was in charge of the whole end-to-end process. And since nobody was in charge of it, nobody paid attention to it. --- **The structural innovation is the process owner.** Unlike functional managers who supervise people within departments, a process owner owns the design of an end-to-end process across the whole organisation. They have sole authority to sanction changes, exercise influence more than control, and need to be an unusual blend of engineer and salesperson. The role is closer to architect than firefighter: the process owner makes sure the fire doesn't start in the first place. What gives the role teeth is enforcement. Hammer recommends giving process owners exclusive control of IT expenditures. Only a process owner can request system development. That puts real constraints on managers who might otherwise ignore the process design and build their own workarounds. A second mechanism: the process owner specifies the performance metrics that operating units must use. When everyone in the old model was measured on something different, no one was measured on anything that mattered to the customer. The new metrics have line of sight: perfect shipment rate, on-time delivery, things every person in the company can see how their actions affect. Rewarding people on a metric like perfect shipment is far more effective than rewarding on corporate profitability, because profitability is too abstract for most people to connect to their daily work. Measurement drives behaviour, and most traditional measurement drives the wrong behaviour. Heroes, in Hammer's view, are symptoms that the process isn't working. A culture of heroism means work is routinely getting rescued rather than designed properly. This is [[Designing the organisation]] backwards, letting structure dictate outcomes rather than outcomes dictate structure. --- **Behaviour comes first; belief follows.** One of social psychology's most enduring lessons is that attitude change lags behaviour change. You don't need people to believe in teamwork before you implement processes that force collaboration. Implement the processes and belief follows. If you compel people to work in teams, they begin to believe teamwork matters. If you push decisions down to the front lines, people start to take personal responsibility. This is the operational answer to [[Psycho-Logic]]: don't argue for change, design structures that make the new behaviour the path of least resistance. The sequencing point that cuts deepest: technology follows process, not the other way around. Automating a bad process produces bad results faster. This is why [[Process power]] compounds so slowly and so durably: it's built layer by layer through operational discipline, not through software purchases. Get the order right and you can do a great deal with modest technology. Get it backwards and no technology budget will save you. Under the old regime, the company Hammer studied met delivery promises less than 15% of the time. With the redesigned process, it hit the target more than 90% of the time. The tasks were largely the same. What changed was how they fit together into a whole. ---