14 July 2014
The other day Martin Read was posing the question of KPIs in the FM World Think Tank. This is a topic dear to my heart as previous columns here show, and I am firmly of the opinion that there is too much made of measurement in general and KPIs in particular.
You have to have some form of measurement in order to be able to manage, but how much do you need? Think about time because that is something that is important to us all in both business and personal situations, but look at how vague we are about it. Yes, some people are obsessive, but most of us talk about time in a very imprecise manner; it's just coming to half-past, I'll be about five minutes, it's just gone a quarter to, and so on. You might time your morning run to the thousandth of a second, but most of the time we manage perfectly well without measuring time to any real degree of accuracy.
Another example of loose management is in driving, where we manage pretty well without worrying too much about measurement. Most of us don't glance at the speedometer often enough and in any case we can't afford to take the time to look for long enough to be certain of whether it says thirty-six, seven or eight. They are not the most accurate of instruments anyway, and their performance is further degraded by tyre pressure and wear among other things. And then there are the fuel and temperature gauges, which just give you a rough guide rather than an accurate reading. Then there is our driving accuracy, but perhaps that is a topic best left alone.
You can find examples in everything you do where we manage perfectly well without accurate measures, so why are we so obsessed with measuring in business? Computer systems mean that much of what we do is measured just by us doing the job and that is fine because there is no extra effort involved in obtaining the data and the first rule of measurement is that the benefit should outweigh the cost so data for nothing is great. If we have to make any effort to measure, then you have to ask how much the measurement costs.
The big problem with data is the time you spend interpreting it because raw data is not much help. That is why your car's fuel gauge tells you that you are at half a tank rather that at 7.652 litres. To manage your driving for most of the time you don't need to know exactly how much fuel you have. If you understand how to use it (and many don't) you can stick out a finger and tap the button to cycle through your trip computer to see how far the car thinks it can go before you need to refill, but the fuel gauge - vague as it is - tells you enough.
One of the main causes of over-measurement in business is lack of confidence and we measure to protect ourselves, to justify our position against another department or a supplier or customer instead of measuring to enable competitive advantage or improvement. Analysing numbers to make the future better is a sound investment, but doing it to defend the past is a waste - and that is why I am against overuse of KPIs.
Yes, we need to know how we are performing, so having the right performance indicators is important. But it is worth thinking about how many of those should be the key ones. Lapsing into management-speak, sometimes we call KPIs cockpit dials so - going back to the car - think about your dashboard. You have two big dials, a couple of smaller ones and some lights, one of the latter being a red one that says STOP.
Take that across to business and five or six KPIs are all that you really need. If you want detail you can find it by drilling down into the base data or in the general performance measures. The other important thing for business KPIs is that they don't have to be the same forever. It is much more sensible to change two or three of them from time to time.
KPIs are a tool and, like any tool, they can be dangerous if not used well. Use them sparingly and understand what you are using them for and they will serve you well.
John Bowen is an FM consultant