A seasoned systems thinker might argue that ideally the work would be cleaned at the front end—in central claims. But first, everyone in the business had to unlearn the old thinking and methods and learn to think and work in the new way. The intervention was designed so that branch staff should participate in the learning experience and act their way into the new way of thinking. This made it easier to implement subsequent improvements. Having made a step-change improvement across the organization, work began to identify causes of variation (things that make time go longer in this case) within and across branches, as well as in the centre. As they were identified and acted on and as people and the organization learned, time shortened, productivity improved and morale rose. The eventual result was a fourfold increase in productivity from the same resources. None of this could have happened without the initial process of discovery. The previous management approach would never have specified the eventual outcome as a target. Their mental model, of 95% targets achieved, everyone working to procedures and so on, could never conceive of such a significant improvement without extra staff. Managers first had to learn how to ‘see’ their presenting problem (unhappy administrators) in a systems framework. Then managers and staff had to learn together how the current system worked (or did not work), and how to replace it with a better one, using the feedback it generated in a continuous process of improvement