“Targets keep you blind to the true performance of your service”
There can be few things that provoke more emotive responses in public sector workers than the subject of targets. Centrally imposed targets have been cited as underlying factors in many of the recent major public sector scandals including MRSA infection rates, Staffordshire Hospital and Baby P. Recently, some senior local authority managers, including some who are members of SOLACE, have released a report (The Illusion of Control) providing evidence of the damage that is being done by targets in their services. Perhaps an indication that targets are falling out of favour (for the time being at least) is the quiet withdrawal of an Audit Commission report, Targets in the Public Sector, from open download. The reality is that management by targets was always going to run into trouble. Wherever there are targets, there are a series of unintended consequences, many of which hit our headlines on a daily basis. From a systems thinking perspective, targets are actually tools and have long been in the frame for damaging improvement, preventing learning and sowing disorder. Here’s why.
To discuss targets invites a return to definitions. What is a target? There are two relevant answers. A target is (i) something to aim for and (ii) a tool, something that helps you to get a particular activity done. The main reason that is commonly provided for the use of targets are that they can ‘motivate’ people to improve performance because there is something to aim for. The evidence however is that targets do motivate people. It just isn’t in the ways that target-setters expect. They motivate people to do the wrong thing.
A systems perspective of performance
Often information is provided to senior managers and leaders in the form of averages (weekly, monthly, annual). This average regularly comes with a target handed-down for workers to achieve. Performance is reported against the set target. If the average sustains and doesn’t drop, or is close to the target then many managers express satisfaction. A different way to display performance is over time. Each x on the chart below (figure 1) represents an abstraction of an individual instance of service provided. It is often the time taken to provide a service (accept or reject an application for example). The assumption made is that the x that you measure matters to your customers.
Figure 1. performance over time shown with upper levels of performance and lower levels of performance
Each x relates to a team or individual’s performance over a period of time which could be a day, a month or even a year. Over time these form a pattern. It shows the most that can be achieved over time (the upper limit), and it shows the lowest that can be achieved over time (the lower limit). When you investigate specific points in the range, it is necessary to understand what causes some to take longer than others. In almost every case the things that influence why it takes less or more time are not in control of the individual worker. It is something that is a product of the system: for example, it could be their training, equipment, the policies and procedures and the complexity of the call, etc.
The systems thinkers at Vanguard have consistently found that the majority of performance is caused by these system conditions rather than an individual’s effort. Roughly this can be apportioned at 95% being down to system conditions and only 5% to elements of performance attributable to the individual. This means that the majority of things that impact the length and quality of a service interaction are beyond the control of those working in the system. Set the target too low and it is easy to meet and the service doesn’t improve. When targets are set beyond the capability of the system, and the systems conditions are beyond the control of the workers, there are only a number of ways to meet the target.
Figure 2. Targets are often set beyond the capability of the system
One common reaction is to cheat to meet the target. How else can workers get better or quicker? They are caught between the target and the capability of the system. In some public services the distress this can cause is extreme and the only way to stop the stress is to do the right thing by the customer and cheat the targets. Managers are not to blame as they are themselves subject to the same systems conditions.
2. Tell the truth
The other option is to be honest and face the regulators, allowing your performance to be published in the league tables. Money and reputation are linked to target attainment. The public sector is scattered with the careers of workers and managers who took this option.
3. Chase the target but do the wrong thing
In many systems, Vanguard has found that the target has worked to the extent that people have focused upon meeting the target. Unfortunately this has been to the detriment of the customer. The presumption is that the performance indicators handed down the hierarchy (which become de facto targets) are measuring the right things. Jeremy Cox, Vanguard Public Sector lead and systems thinking expert argues that the ‘vast majority of performance indicators are measuring the wrong things’. The impact of adding a target means that the wrong focus is driven in deeper at detriment to the real purpose of the service. The examples are legion. The 8-week target for local authority planning applications is one. It has created a planning system that is focused upon making defensible decisions rather than saying yes to good developments and helping people to submit good designs. When you look at the end-to-end times for planning applications, at or near the 8 week target, the number of application being accepted or rejected rockets. Has the target improved the experience of those submitting planning applications and has it helped them to create better developments? No. But the target has been met.
Figure 3. Planning applications accepted or rejected spike to meet the target
Another example is from housing repairs. The target here states that emergency repairs should be carried out within 24 hours, 7 working days if it is deemed to be urgent and 28 working days if it is a routine repair. This sounds plausible and a large majority of housing organisations score highly against the target. The reality of performance from the customer’s perspective is very different. When organisations study their services as a system, they understand that the true purpose of the system is to fix the problem in as few visits as possible. When the organisation begins to measure true end-to-end time (from the customer’s point of view), they discover that it can take as long as 220 days! It is possible to meet the target and yet it can take up to 10 visits to fix a problem. Each report is counted as an individual job. To meet the target, the result may be a patch repair instead of a permanent repair. Target met? Tick. Customer experience? Terrible.
Does this mean that using the correct measures, and then using a target would lead to a better outcome? Well no. It is because all targets (and standards, etc) are arbitrary. They are a figure plucked-out of the air. Even workers setting themselves targets do not know the true potential for improvement in a system. Results that organisations working with Vanguard have achieved have been well beyond of any targets set by leaders or government. In some instances, there have been improvements of thousands of percent, in most cases improvements of hundreds of percent.
Vanguard has many examples now and they all tell the same story: people’s efforts are focused on the target at the cost of doing the right thing. Jeremy Cox argues that one of the most revealing things about targets is that you can’t see the sub-optimisation they cause because they have created the illusion of good performance. Targets keep you blind to the true performance of your service.
Ignore Targets and Change the System
Systems thinkers don’t reject targets outright. Instead they believe that perfection is the only target worth aiming for. This turns the whole concept on its head. All work is based upon knowledge and experimentation. Decisions about how the work should be done are placed back into the hands of the workers. The job of a manager becomes to remove the things that stand in the way of workers doing a good job. Doing the right thing is delivering what matters most to customers. Perfection is beyond the capability of the system, but progress towards it continues to be an aim, ensuring continual experimentation and improvement. Leaders spend time in the work, understanding the measures being used and gaining knowledge about how the system works. Vanguard argues that the true purpose of measures is to learn and improve. And targets don’t help learning.
Reconnecting leaders with their organisations
Jeremy Cox highlights the problem with a question. ‘If we know that targets damage service and the evidence keeps piling-up’ he asks ‘why try targets harder?’ He is referring to SMART targets (the acronym for Specific, Measurable, Achievable, Realistic and Time-constrained) and the newer politically-floated conception of targets as entitlements. Perhaps what this brings us back to is the fact that if leaders want to be effective, they must be able to accept that the command and control paradigm never really delivered. It means that leaders and their services were always dislocated from the work, and only ever connected by an illusion of control. The figures that the machine churned out told them great stories of success, and as they passed-down ever-increasing volumes of targets, their service users suffered and the staff just walked away.