Home > Musings > Measuring Performance in Public Agencies

Measuring Performance in Public Agencies

July 13, 2011

The dictionary defines Performance as ‘an action, task or operation, seen in terms of how successfully it was performed’. Recently the Chief Minister of Karnataka, in his bid to have an image makeover, announced that he would regularly review the performance of the different districts and the State’s development efforts. Being interested in the subject of performance in public agencies and having some expertise in it, I saw his move as a welcome development, only to be sadly disappointed. He visited Mysore and in a span of less than 60 minutes completed reviewing the performance of all Government Departments in the district. I was amazed at how this was done and spoke to a few senior district-level heads of the departments. I soon realized that this was less of a review and more of a PR process. During the course of the conversations, I also unearthed that most public servants have a rudimentary understanding of ‘Performance’ and how it can be measured.

Most of us have a tacit knowledge of Performance. We can recognize and understand that something is indeed working as it should be and from a very young age learn to appreciate performance and quality. How does one translate this tacit understanding into something more structured and measurable? Can one actually measure Performance in the public sector? One can learn a lot from the private sector that has taken the lead in measuring performances of individuals, teams and entire organizations. We also need to understand that the public sector is vastly different in the context in which it operates and has more egalitarian objectives, but isn’t this the very reason why its performance should be known and measured? By virtue of being the public sector, it needs to be more accountable and transparent but paradoxically one does not find these as the primary drivers of performance. By the very nature of having unlimited resources at its disposal, the public sector also tends to become inefficient and opaque over time. Measuring performance in such a complex scenario needs knowledge of the program being measured, competent people to do the measuring, understanding of the tools deployed and the metrics of measurement and more importantly the context – before one indulges in this activity. What I have tried to present is a very simplistic beginner’s version of doing the same over 5 sequential steps.

1. Define performance

I would rate this as the most critical step. One can measure performance only after one defines what it means to us as individuals, as teams and as an organization. This definition has to be context specific and should flow from the understanding of the organization’s mission, vision and key objectives. We need to realize that most organizations manage only what they measure and a clear and proper definition of Performance will actually drive the organization to a better and a more responsive management. Performance can mean different things to different people. It also varies on whether one is looking at the performance of the system from the point of view of inputs, process, activities, or the outputs. People tend to take the easy way out and limit themselves to looking at performance at the output level that may indeed help measuring effectiveness but may not capture the efficiency of the system or the processes that are driving the system. A good definition of Performance also helps set the agenda for all the players involved in the system and one can then use it as a tool for both monitoring & evaluation, and for future planning. One of the key challenges for public agencies is to benchmark the services that they provide to the citizenry. Citizens have their own perceptions of what their governments should be doing and the expectations also vary from place to place and from department to department. On an everyday basis, we would normally not worry too much about the inefficiency of the military or the country’s external affairs ministry but would surely be affected if the postal department or the public transport department functioned poorly. Keeping sight of the extent of citizen engagement and their sense of entitlements will help design a good performance system and one can then look at the acceptable bench markers that will be both practical and aspirational. Many public agencies are today declaring ‘Citizen Charters’ and ideally these should form a part of that agency’s Performance definition.

2. Decide on the metrics

Defining ‘Performance’ will be critical in understanding how to measure it too. The metrics need to capture comprehensively and in a simple framework, this definition of ‘Performance’. Expectedly the metrics will vary based on the definition and what is important for the organization to be measured. Measurements need to be inexpensive, simple, user-friendly and developed as a bottom-up exercise in a participatory manner. The scales need to be context-sensitive and should take into account the competence of the people using the instrument. The common mistake of using metrics that is convenient rather than that which is essential is to be avoided. Most organizations stagnate after developing the metrics, but generally do not see the need for a constant revalidation of the same. Metrics will also evolve, change and refine as the context and the focus of the organization changes. It is best to begin with simple metrics and build incrementally as the organization and its personnel mature. Many performance processes fail not because the metrics are poor or the process weak, but more because of the complexity of the metric, the difficulty in applying them and incompetence of the people doing the measurement. One must also understand the insecurity and HR disruptions that measuring performance will create and the people need to be sensitized and gradually built to adapt to the demands of such evolving systems.

3. Determine who will do the measurement and when

Most organizations find it difficult to have a dedicated person or team to oversee the process of performance measurement. Some of them are keen on developing a system and then expect the system to somehow ensure that performance will occur. Performance measurement should be led from the top. The whole organization should know and understand that it is not just a new fad that has been introduced, but a serious process of bettering oneself. For best results, one would expect the head of the agency or the CEO to take on this responsibility and ensure that all audit and information processes constantly feed information and analysis directly to him with little or no interpretations. One other danger that organizations need to be cautious about is the periodicity of undertaking these measurements. Periodicity is driven again by what is being measured. Some critical processes may need constant monitoring; some inputs may need to be measured only when the cost goes above a certain critical amount; outputs may be time sensitive and has to be measured only after the appropriate time needed for it to occur has lapsed between two measurements. Outcomes may need to be measured and evaluated only once in a few years after one is sure that sufficient time has allowed for the impact to be seen, measured and reported. Cost considerations should also be kept in mind when the periodicity is planned. Performance processes should not end up being costlier that what is being measured.

4. Build a review process

Demanding performance of oneself and an organization is a very exacting process that requires discipline, determination and a strong political will to undertake. It requires not just managerial knowledge but a visionary leadership that is constantly evaluating, refining and improvising processes all the time. Performance processes normally fail because the required discipline and rigor wanes over time – one must have the patience and the perseverance to allow the system to mature for results to be produced. Measuring performance is not like instant coffee – made quickly and giving immediate gratification. These systems take time to initiate, evolve, mature, and become organizational culture. Leadership needs to be constant, consistent and serious till the entire cycle has taken root. It also needs mentoring support from experts who are willing to not only design a review process but also facilitate its implementation in the initial phases. The core leadership should take it as sacred responsibility and be willing to make public disclosure of achievements or variances. Success or failure should not be determined merely by the variances that occur and one must be willing to see it as ‘work in progress’ for most of the time. One must have public displays of the review process and all stakeholders should have a say in not just the design but also in the actual framing and implementation of the reviews. Reviews should be seen as sacred events that would not be moved about to suit the convenience of any person or context. They should be time sensitive and many a review is known to lose its direction because of a weak convener or chairperson running the event. Normally reviews are to be chaired by the head of the agency after suitable preparation. All data pertaining to the review should be prepared and shared in advance to have meaningful discussions rather than ritualistic ‘shows’. Apart from the time, date and other operational logistics, determining who should or should not participate in these events is also critical. Participants should not be determined by the formal or informal power one holds or for politically correct reasons, but more by the value that a person adds to the entire process.

5. Match measurement with a follow-thru process

Most reviews tend to be successful only when one walks the talk. Reward and punishment behaviors are indeed critical for human performance and public agencies should move away from not wanting to indulge in them. There is a normative feeling that public jobs are sort of permanent and career growth is not necessarily dependent on performance. One must communicate that ‘mediocrity’ need not necessarily be synonymous with ‘public agencies’ and a culture of valuing performance should be created. This can be done only when good performance is rewarded and poor one punished. A simple tool could be naming both good performers and poor performers. Peer opinions and grapevine discussions serve to motivate enormously and one should capitalize on naming and shaming processes in a very titrated and dignified way. These processes could also become counter-productive and only the top leadership of the organization should be permitted to indulge in them.

Performance is usually seen as another ‘management’ process to move the bottom line. Performance should not be part of the ‘command and control’ toolbox and one must go beyond this and see it as a constant ability to inspire, drive and elevate human potential. Performance Leadership is indeed one of the most critical elements in the entire process and forms the very platform on which all further processes and systems have to be built. What will differentiate good from mediocre agencies is the extent to which one is willing to take the challenge of ‘Performance’ and measuring it head on. Nations like India can ill-afford inefficiency and Performance leadership could be critical in ensuring that public agencies not only deliver on their mandate, but also become role models for others to emulate.


Categories: Musings
  1. July 19, 2011 at 8:31 am

    Thanks for this nice post, Balu.

    For a person who himself does not perform, it is difficult to objectively measure others’ performance. Many of the performance measurements and appraisals are mockeries in public agencies as well as in private organisations.

  2. Arun Karpur
    July 14, 2011 at 2:00 am

    Balu: This is an exciting post. I appreciate how you have made such a technical issue readable for common people. I would like to add few points to your discourse on performance assessment in public sector organizations.

    As businesses, the public sector organizations must strive to transform itself into Learning Organization – one that has built in organizational structure and processes to capture learning, synthesize observations and build or refine activities for achieving better outcomes. It is critical that one view this as a continuous process of quality improvement. More often public programs subject themselves with snapshot evaluations without any feedback loops – a necessary element of the change engine in a systems improvement process.

    Additionally, performance assessment is not all about just the metrics as more often metrics tend to define outcomes and many times agencies fail to recognize the difference between process outcomes and program outcomes. So I suggest agencies also consider looking at processes undercutting their activities with an intention of improving efficiency as well as effectiveness.

    Also, in my view, often performance assessments ( I personally like program evaluation better) in our subcontinent is always driven by the interest to collect data on large number of individuals, which I believe is not necessary. Efforts should be directed for learning from subjective experiences or qualitative data to build the depth necessary for a critical reflective synthesis providing a holistic view to performance. India’s health system is a prime example where targets are set without understanding as to how one would reach those targets and if those targets mean anything from societal or systems perspective.

    One complicating side of the public sector organizations is that most are constrained by the political environment and their mission is twisted (or tweaked – a more euphemistic expression) to suit the political demands. From that perspective, the public sector organizations must build an extensive network of internal as well as external resources/think-tanks to help them remain to be creative to be effective and responsive to the political demands imposed upon them.

    Finally, I believe performance assessment of public organizations is as much a technical exercise as it is of leadership. Both are necessary and required elements of a strong service system.

    Arun Karpur

  1. No trackbacks yet.
Comments are closed.