Measuring human performance by using an annual basket of ‘objective’ parameters can most often yield sub-optimal results. This is primarily because old yardsticks no longer hold good, either in terms of the metrics that are applied to gauge individual accomplishments, or in terms of defining what the collective goals are. It is not surprising, therefore, that large organisations such as Accenture or Adobe have begun moving to a more dynamic ongoing method of reviewing employee performance.
There are compelling reasons to make the shift. Looking back at the past year to evaluate a worker’s achievements, inevitably results in a system of ranking and ratings. Statistically this can be fallacious. It can result in measuring one person’s skills against another, although the goals for both may be different.
This is truer in highly diversified services companies where performances are intangible variables as opposed to an assembly line factory floor. The so-called ‘Bell Curve’, the metric that has assumed a Holy Grail status in the world of human resource management, can also lead to flawed results. For instance, it is not useful in a small population and has a tendency to underrate top performers and overrate laggards.
In an inter-connected world, an employee’s achievements are linked to the performance of her colleagues. The ‘annual performance review’ system masks an employee’s contribution, or the lack of it, to the organisation’s overall vision and goals. This is the reason why many surveys and experts point out why annual rating-based reviews discourage teamwork and collaboration.
As an alternative, a regular ‘review-as-we-go-along’ model is always a better determinant in appraising an employee’s performance. The doctrine of preview is predicated upon goals and milestones that are set in consultation with the supervisor. With expectations set in the beginning and linked to the manager’s targets, the employee should be encouraged to lead the discussions during the periodic evaluations in an informal way. One may argue that this is not a perfect structure, but the gains from a preview method far outweigh the damages caused by an opaque rear-view method.