Complexity breeds Complexity
The need to deliver varied and flexible services is just one reason why organisations now operate more sophisticated processes. This results in the thirst for more complex data. While simple processes can be managed with simple data, more complex processes demand detailed data. Internal and external stakeholders often want different information and the use of multiple systems within an organisation compounds things further. Lean encourages process simplification, proportional to the purpose and desired outputs of the service – removing non value-add activity resulting in a simpler process. The same principle applies to handling and using data.
Why is complex data a problem?
Data is clearly important; merciless measurement is a fundamental principle of Lean Thinking. It enables organisations to understand their services and decide how to improve. The volume and detail of the data now available makes this difficult and often leads to clouded understanding. Common issues include:
• Organisations adopt a ‘just-in-case’ approach to data capture and reporting
• Organisations are data-rich but business intelligence poor
• Different stakeholders requesting different data, or variations
• Systems set to record everything rather than only required information
• Different data stored on different systems, often incompatible
• Inflexible reporting tools for data extraction, which don’t meet needs
• Overly-complex data leading to errors and inaccuracy.
What does good look like?
Moving to a more effective approach means rationalising the data set used and trimming back to only useful data. We often state: “do not collect data that is not recorded, do not record data that is not analysed and do not analyse data that is not used to make improvement”. Any data not used to improve performance is waste and time spent pulling it together is non value-adding activity. While the data required may still be complex, it should be no more so than needed.
Getting your approach to data right
Gathering and using data proportionate to the nature of your service is crucial. This means approaching data differently.
1. Firstly, define your service’s purpose, followed by defining your process(es), inputs, demand, capacity, outputs, outcomes and problems
2. This enables development of a solid set of performance measures, from which a data set can be compiled. Developing the right measures – simple, leading information about performance – results in simpler data sets
3. Making operational teams responsible for collecting and using leading performance data will drive simpler data sets – avoiding the need for complex system-driven reports
4. Stop measuring things that are not useful and stop collecting the data
5. Manage the team’s changing requirements by asking, “what is the purpose of this measure and data?”
6. Standardise and automate any data sets that are business essential non value-adding.
