So far, I’ve focused on problems. This is about solutions. I have tried to ensure that the solutions I recommend are conceptually and scientifically robust, but also practical and easy to apply. Modern performance analysis must be:
Dynamic – to expose trends
Performance is a pattern of behaviour that cannot be captured by focussing on individual data points. We need to measure and analyse performance in a way that exposes trends and trajectories.
Filtered – to deal with noise
We must be able to separate the wheat (the signal) from the chaff (the noise) to focus our limited attention, knowledge and intellect on the right things and avoid being distracted by – or worse reacting to – randomness.
Simple arithmetic can’t do this for us. We need to view data through a probabilistic lens and use statistical filters to help us extract insights from big noisy data sets.
Reframe performance – to reflect its complex multi-dimension
We need nuanced targets or comparators that are less arbitrary than the targets produced by traditional processes like annual budgeting and better reflect the actual business context and performance potential. They have to help us track performance over time, rather than being anchored on points in time, and help us filter out the influence of noise.
Communicated effectively – to exploit the capability of our brains
It is imperative for us to develop a better understanding of how to communicate meaning effectively to an audience of decision-makers with differing knowledge, experiences and capabilities. This needs to be based on an understanding of how the brain processes information and a grasp of good design principles that exploit its capabilities.
I have argued that we should not look to technology as a silver bullet that will solve all our problems for us. But it is equally clear to me that we will not be successful in achieving any of these goals without the help of technology.
However, we first need to learn how to best exploit it. All of the techniques and approaches I describe in my book can be deployed using paper and personal calculators if you want to adopt a ‘back to nature’ approach to performance measurement.
This would be perverse, so I expect and encourage you to experiment using desktop productivity software such as spreadsheets. Excel will, however, only get you so far. To implement methods that I recommend at scale and at speed in real life you will most likely need specialized software – but having experimented using simpler, more forgiving technology you will have developed a good grasp of what will work best in your organization and what you want from a specialized software solution.
I have also explained that I do not believe that dashboards are the solution to the problems of communicating meaning – primarily because they are designed to support personal understanding and enquiry not the social process of decision-making.
Enormous progress has, however, been made in the last few years in understanding and codifying good practice in dashboard design and graphical communication, led by pioneers like Edward Tufte and Stephen Few. And these principles can be applied directly to the design of performance reports even if they continue to be produced in Excel or PowerPoint and distributed on paper.
Again, I encourage you to practice with these ideas and experience for yourself the impact they can make. And what you learn can then be applied to customizing the dashboard software you have already got or are minded to buy. That’s enough preamble – let’s start ambling.
This is an extract from Steve Morlidge’s latest book, Present Sense.