A big elephant in the room
McKinsey has OGSMT, Deloitte calls it integrated performance management, and at Capgemini Consulting we call it Line of Sight instead of Performance Measurement. No matter how or what you call it, it comes down to have a set of clearly defined metrics that are related to one another so that each level in the organization is able to manage its performance in line with corporate strategy. Whether it's the top-execs or operational managers, if KPIs aren't aligned chances are that people aren’t on the same page and headed toward different goals on the horizon. So how come we find that KPIs are misaligned? And what is the best practice in defining holistic KPIs? Obviously there's a big elephant in the room, everyone nowadays has a set of KPIs that measures some kind of performance. That's not what performance management is about. It's about creating insights from those KPIs and how benchmarking the results are meaningful in the first place. All too often I've come across KPI dashboards that lack this coherent view, or only tell part of the story. Stripping Performance Measurement down to its essence, there are six principles that govern a good set of KPIs.
Six principles to KPIs
- It's the magical Number Seven. A proper dashboard contains roughly about seven items, not much more, not much less either. Seven items is about what a human brain can effectively handle to process and remember.
Dashboards should be objective and relevant. Deming has put forward his DMAIC cycle which helps to objectively measure process performance. First Define what and how you're going to measure, then Measure it, Analyze it, Improve it, and finally Control it. And then it starts all over again in order to improve the quality and relevance of your dashboards.
Improving overall performance requires managers to focus on a lot of things at the same time. Back in the 1990s Kaplan and Norton introduced their Balanced Scorecard, which helps you not to focus too much on just one aspect, but to - yes, indeed - balance the overall performance. Cascaded down, KPIs consist of Perfomance Indicators (PIs). Focus too much on just one PI might result into too little impact on the top-level and the ever so much desired improvement will be achieved too late. Kaplan and Norton summarized it quite right with "If you can’t measure it, you can’t manage it. If you can’t manage it, you can’t improve it."
- There's the factor time to take into account. The Six Sigma field introduced the notion of Control Limits which acts as a handy panic-button. When performance stays within the control limits there's nothing to worry about. When outside... well you know where I'm heading.
- An interdependence exists between your KPIs and PIs, and the overall behavior is driven by the settings of your system. System Dynamics teaches us the importance of recognizing at least two things, (1) there's a difference between indicators that act like stocks or like flows, and (2) a system's behavior is driven by the stocks, not the flows. In Supply Chain terms, we're always looking at sales of churn (flow)rates, but all too often we neglect the actual inventory or installed base levels (stock). Being able to recognize a KPI as being a stock or flow-indicator helps you to improve performance effective and efficiently.
Yet, you're not there by mastering these five principles, because it doesn't tell anything about the performance compared to another. The last principle is about standardization. In order to compare you KPIs and PIs with your peers or even-cross industry it is necessary that the definitions of your KPIs (remember DMAIC) are nicely aligned. There's a plethora of frameworks and process stacks available that can help you with setting up a standardized KPI and process framework. Yet, for Supply Chain there's only one and that is the Supply Chain Operations Reference (SCOR).
Don't reinvent the wheel
If your current dashboard and reporting structure is not yet in line with these principles or you need to start greenfield, it would be wise to define your KPIs and relating processes as vanilla as possible. Start off with the best practice process stacks relevant to your industry and customize where it is absolutely neccessary. Consequently, validate the designs with the relevant stakeholders from top to bottom in order to achieve line of sight. Or put differently, align the corporate strategy to your operational targets. In terms of manageablitity, don't forget to attribute weight to the most important KPIs while keeping an eye out for the most important drivers. Obviously, some IT tooling will come in handy to speed up the reporting process and to increase data reliability