I was talking to a client about KPIs and they passed me a large bound booklet. It must have been 200 pages. The booklet was their operations report; a compendium of KPIs. I was asked if I could assist to determine new KPIs to address performance issues within the business. I flicked through the booklet. It was filled with red and green status indicators. Mostly red. I handed the booklet back to the client and apologised – I could not help him. I pointed out that the book already held more KPIs than I could possibly think of, and that the problem was not the absence of a specific KPI, but rather the absence of response to the existing KPIs.

To elaborate this point I turned to a page at random. The three monthly trend showed three months of ‘red’. This was largely the same for all the indicators on the page. The KPI’s had been indicating an out of tolerance position for months and nothing had been done about it.

The conversation then moved to discussing the purpose of KPIs and how to make them effective in the business.

Frequently when first implementing KPI’s, the mindset is to measure everything that moves; if it doesn’t move, kick it until it does and then measure it. Unfortunately, within a couple of months, many KPIs will be found to offer so little value that they are ignored. This sends a message to managers – that it’s all talk and no action – and the slide to indifference starts.

There is only one reason to implement KPIs and that is to manage behaviour. Equally there is only one type of behaviour in an organisation and that is human behaviour. Companies don’t behave, technology does not behave, plant and equipment does not behave. People behave. But what do KPIs measure. They measure customer satisfaction, mileage on trucks, lost sales, revenues, costs, throughput, quality and productivity. The list is endless.

The underlying principle of a KPI is – if you are going to manage it, then measure it, otherwise ignore it. There is no point in measuring something you are not going to manage. To address an indicator that is out of tolerance somebody somewhere in the organisation has to do something differently. Different to what they did last time because last time they created an out of tolerance result.

The difficulty arises when the KPIs are too broad in their definition. For example; sales are down so salespeople are told to ‘get out there’ and work harder. But does the salesperson know what to do differently to improve their sales figures. Clearly their current approach is not working. They need to ask themselves – what must I do differently to ensure I get a different result – or in other words how must I change my behaviour.

The astute sales manager will know exactly what behaviours are required to make even an inept salesperson successful and they will implement KPIs that evaluate this behaviour at the micro level. The focus will be on things that the manager can actually change. This could be number of calls made, length of calls, number of times the salesperson gets through to the decision maker, appointments set and the number of times a salespersons telephone rings.

The misconception is that sales managers are in sales, or more broadly, that managers of a function are in that function. Truth is they are not. They are in a separate function called management and managers behave differently to workers (staff). Therefore the KPIs that measure a managers’ behaviour must be different to those that measure staff behaviour – even if they appear to be in the same function.

The seniority between managers introduces additional complexities.

Consider the following four relationships.

The first relationship is between first line management and staff. A key feature of this relationship is that the two roles are different. Broadly, staff deliver technical output while managers deliver an administrative output. This requires that the manager use KPIs that measure ‘technical behaviour‘.

The second relationship is between a manager who manages workers and his/her manager. 

 The next level is the relationship between two levels of management where neither level is managing workers.

The key feature of the previous two levels is that management is managing management. This requires the use of KPI’s that measure ‘management behaviour‘. The KPIs for each level can largely be the same, separated by the degree of aggregation applied to the more senior of the roles.

The fourth level is the relationship between directors and managers. In many respects this relationship is similar to the first level of management. From a directors point of view, business management is ‘technical delivery’ whilst directors take more of a deterministic/directive role in the organisation in that they determine strategy, the governance model, the organisational risk appetite etc.

For each relationship set, what is the role of the senior manager. This is a complex question and not easily answered in a short paragraph. But in the spirit of this article I will say; it is to ensure that direct reports are behaving in a manner that maximises the likelihood of them achieving their KPIs.

Based on this definition, for each relationship, it is mandatory that the senior of the two managers knows what behaviour is required to maximise the business benefit possible from the subordinate management function, and what behavioural changes are required when the expected business benefit is not realised.

If the relationship between the KPI and behaviour is not understood or if a KPI cannot be manipulated through changes in a managers behaviour then it is likely that the wrong KPI is been used.