An Appraisal of Appraisal Systems

In a certain discussion forum, a number of brilliant people have expressed views about the positives and negatives of appraisal systems used in the industry and other settings for assessing the performance of employees. The discussion inspired me to review the whole issue in detail in the form of an article. 

All appraisal systems in a sense take away the initiative and the power of decision from those whose work is subjected to appraisal and, therefore, gradually re-shape the thinking-style and working of the organisation in a direction that accords with the aims, objectives and priorities of those whose vision permeates the appraisal system. If there is no such vision behind the appraisal system, then the appraisal system itself must be taken as inadequate. The appraisal system, if it is to serve its purpose, must carry a clear, though implicit, message to the employee: "This is what I expect of you and my assessment of you will depend upon the degree to which you perform to achieve these objectives."

In so far as an appraisal system tries to set the direction of the organisation and then assesses the employees' performance in terms of his contribution in moving the organisation in that direction, there cannot be much disagreement.  The problem arises when the employee feels that the appraisal system is in fact ignoring a lot that he has done to further the objectives of the organisation. In so far as the objectives are concerned, the employee cannot have a quarrel, because it is the owner's prerogative to set the objectives. The disagreement arises about whether the appraisal system succeeds in measuring all that the employee has done to achieve these objectives or fails to measure some of his efforts.

To give an example, a work-extractor might produce better immediate results in statistical terms but may ruin the atmosphere of the organisation. Another person may be slow, but may bring about fundamental changes with far-reaching future implications. The question, therefore, is whether the appraisal system is tuned to measuring the work of this slow-and-steady. Unfortunately, most builders of appraisal systems, in their  eagerness for accuracy make it statistics-based and thus exclude all those parameters that may require impressionistic evaluation.

The disconnect between what is sought to be measured and what the appraisal system actually ends up measuring is the crux of the problem. This problem arises because the agency that prepares the appraisal system is either not aware of the long-term and short-term factors that impact the attainment of objectives in that organisation and the very subtle manner in which this may happen --- to put it another way, has an externaility of approach ---or is convinced that the criteria chosen by them measures exactly whatever is intended to be measured.

Scott Simmerman makes this point beautifully and I quote him::

" The biggest problem with these appraisals, in my humble opinion, is that
they are generally pretty terrible in getting at the ESSENCE of what is
needed to operate successfully. You tend to get people focused more and
more on the numbers rather than the desired behaviors and, over time,
you get exactly what you want (or at least what the appraisal drives). So, the "poor performing" associate who is not handling enough incoming
calls a day (the calls are too long reducing the number of them being
handled) just might be tempted to disconnect on that customer whose
situation requires that some time be spent on that call -- the customer
service rep who gets the "new installation" call and who has to gather
all kinds of information about the customer, their location, their
credit, etc. "Hey, not my fault the call got cut!" if they even might
run the risk of being caught at this. Deming, the quality guru, was adamantly against them for a variety of reasons. In his broad experience, they were miserable, working against quality."
 
Simmerman puts his finger at a very relevant point: statistics-based appraisal systems change the objectives to be pursued. Increasing the number of calls attended by the receptionist becomes the objective rather than giving proper response to the callers.

This problem of the disconnect between what is sought to be measured and what is actually measured and then made the basis of an evaluation exists in all organisations. Even at the national level in a democracy, the public, who are the ultimate owner of the democratic system and should judge the performance of a government holistically, are swayed by a few criteria. In a sense the press and the opposition parties subtly put foward before the public an appraisal system which makes the evaluation partial. Take, for instance, the last general elections in India. The BJP projected itself as a party which, during its five-year rule had made India an "India Shining". The Congress took the BJP on their own ground and argued that the economic growth had benefited only the urban middle class while the rural masses had been ignored. In every election meeting, the Congress leaders gave statistics of farmers who had committed suicide because they could not pay back the loans taken by them. An appraisal system that was handed over to the masses had only a one-point scale: has the BJP ignored the rural masses? Is the BJP, therefore,  a party of the urbanites, by the urbanites and for the urbanites? The press and the media, while giving news-coverage to the election speeches of the Congress leaders indirectly strengthened the validity of this scale. Some may say that there were other issues like the handling of Gujarat riots. But, whatever may be the case, a relatively simple scale of criteria was ready for use by the public. It is the same story in every election in every country.

Apraisal systems suffer from proneness to simplification. The question is how to minimise the simplification. This is where, I think, the modular approach suggested by me becomes more relevant. In the modular approach the organisation is viewed as a module of modules and not as a system of systems. There is a fundamental difference between the two. A system of systems approach takes an analytical view and tries to understand the mutual interactions among the components. A modular approach does not dispassionately view the interaction from a vantage point outside the organisation, but actively and constantly studies it, as one component studying other components from a slightly higher rather than external vantage point, and modifies it. The appraisal system, training, planning and all else become a part of a single module. It is the equivalent of Gandhiji's Panchayati Raj where the Panchayat is a self-sufficient and self-contained module and the country is a module of modules.  

I will be glad to discuss the modular approach with those who wish to find out more about it.