I remember hearing the old sayings about performance measurement early on in my career, and thinking they were pretty astute – catchy and encapsulating a profound truth. Things like:
“If you don’t measure it, you don’t manage it!”
“What gets measured gets done!”
And I’ve always taken them to mean something positive – if you really value your objective, then measure your progress towards it. If you’re not measuring it, it shows you don’t really care. So, if you really do care, and you want it to get done, then measure it, report it, etc.
But recently that last one – “what gets measured gets done” – has been making me think. There’s a negative side to it as well.
The way you drive
The truth of that saying – “what gets measured gets done” – struck me when I drove my son’s Merc once.
The dashboard has a mode to show how economically you are driving. It lights up green in different segments if you’re driving in ways that maximise fuel efficiency – braking, accelerating, steering.
The funny thing was that once I saw that dashboard, I changed the way I drove. I was determined to make the dashboard go as green as possible. And I had a bit of fun trying to make the journey average ‘miles per gallon’ figure go as high as possible.
(If you drive a BMW or Audi, you probably haven’t got a clue what I’m talking about – you only have one measure… “am I in front of everyone else?”! )
The point is that measuring and reporting the fuel efficiency, and the KPIs for the “efficiency drivers”, made me drive differently.
Measurement and reporting affects behaviour
The key thing to understand is that, generally speaking, measuring and reporting something affects behaviour. And I think that’s true even if you don’t position it as a performance indicator, and even if you don’t set a target.
So, hence the saying, “if you don’t measure it, you don’t manage it”. If you want to influence behaviour so that something gets done in your business, then measure it. “What gets measured gets done!”
And that’s normally put positively, in terms of working out what the critical success factors, the value drivers, are, within the business. And then if they truly are critical and important, then measuring them somehow helps to manage them. Even if you start off being uncertain how to influence them one way or the other, when you can measure them you can start to experiment and see the results.
And further, you don’t even have to give incentives or set targets for most things. People seem to have an inbuilt desire to influence things and improve them.
Don’t measure what you don’t want to manage
But the negative is also true. And this is what I’ve come to realise. If you measure the wrong things, you can do real damage to your business.
That’s because measuring the wrong things drives the wrong behaviour. It means people in the business think they’re improving things when they’re actually doing the opposite.
And this is slightly complicated, because this applies to:
- Measuring an outcome that can be influenced/driven in several different ways, some of which cause other undesirable outcomes;
- Measuring something that isn’t important;
- Measures that can be manipulated or misinterpreted;
- Confusing the meaning of a measure – thinking it’s a leading indicator of something it’s not.
So, we need to be careful what we measure. Because, “what gets measured gets done”.
If we measure the wrong things, we may do the wrong things. At best the actions may just waste time, because they’re pointless and don’t matter. At worst the actions can be detrimental to the business.
Why do we report ‘Headcount’?
So, what’s an example? One thing I struggle with is understanding why it’s seen as so important to report ‘headcount’.
It’s probably the first non-financial measure to be added onto a monthly P&L to make it a ‘management report’.
And then we go and compare it to what we had in the plan. Worse, we automatically assume that higher numbers are bad. What’s wrong with having more people? What’s wrong with having more people than we planned?
Someone may say, “but it’s a driver of cost”.
And the reply to that is, “what’s the point putting the headcount figure on the P&L then?” The staffing costs are visible in the P&L anyway. All the costs are in the P&L report. And we know that the activities of people drive cost. So, what additional insight does ‘headcount’ give us?
The number of people working in the business is not as important as what they do in the business, and the critical success factors that they influence.
‘Headcount’ takes no account of mix or relative cost. It also doesn’t take into account business volumes or demand.
And when demand is high, reporting headcount in a way that suggests we ought to be limiting recruitment can put severe constraints on the business. You can just hear the manager saying, “I know all my people are burnt out, 20% of them are off sick with stress and 20% have left in the last year… but increased headcount would have looked bad on the management report.”
As a second example, ask yourself what behaviour is driven when you include cost allocations on a business performance report?
I can tell you exactly, because I saw it in the first business I worked in after leaving public practice. And I’ve seen it many times since.
The behaviour you get is management focusing more on the relative share of cost that is attributed to the different business lines than the absolute amount of the cost. The discussion centres on “fairness” and whether there is a better basis of allocation – could it be revenue? Headcount? Floorspace? Direct cost?
This wastes so much time. My advice is: don’t show allocated costs on business line/unit P&L reports! I wrote about this in another article about Finance items that waste time in management meetings.
Seriously, just take the unit/product/customer P&Ls down to “Contribution” after direct costs, and treat overhead costs separately.
If you have to allocate costs for regulatory or tax purposes, you don’t need to show that to your management colleagues. Why should it matter to them what the regulator or tax office wants to see?
And if you want to do some approximate allocations for pricing analysis and product/customer costings, then fine – kind of. Even there it’s a bit spurious, and can lead to incorrect decisions. The only reason you’d do it is for some kind of cross-check.
Just as an aside, the starkest example of the questionable wisdom of allocating overheads to products, customers or business units, is where a product looks like it makes a loss, because it has an allocation of overheads to it. So, the decision is made to cut the product in order to cut the losses. Lo and behold, the overheads don’t change, and just get allocated to other surviving products, and those become less profitable. And in fact, the overall business is less profitable afterwards, because there is less revenue to cover the overheads.
Or maybe the action in response to a loss-making product could be to increase prices. But that drives demand and volume down, which then makes the product loss-making again.
Show direct costs as direct costs, and overheads as overheads. No allocations!
Absolutes rather than ratios
Finally, one of the things we, as experienced Finance people, should know is that absolute numbers don’t tell the full story. You need to understand the relationships between the numbers to really understand whether good things or bad things are happening.
So, show the ratios instead of the absolute numbers. If the full P&L is misleading, why not be radical and not show it in full. Just show the key numbers and ratios.
For instance, if debtors goes up, that could mean that customers are delaying payment and we need to make more effort chasing them. On the other hand, it could mean that we are just generally selling more. It could mean that we got a big sale right at the end of the month. Or it could mean that we changed some or our payment terms.
Days Sales Outstanding (DSO) is a much better measure.
Be careful what you measure and report
So, be careful what you measure and report. Because, “what gets measured gets done!”
Think about all the figures that you include in all your reports. Think about all the other sources of information that your managers have.
Put yourself in their shoes, and ask yourself how all that information can be interpreted.
Ask yourself what actions can be taken to affect the numbers being reported. Can manipulative, non-value-adding actions have an effect as well?
What behaviours and actions are you expecting or wanting to drive and influence by reporting the measures you report?
Are your KPIs focused on value drivers? And will those KPIs drive the right behaviour, or wasteful or value-destroying behaviour?
“Measure what matters. But be careful what you report!”