The news that global professional services firm Accenture is abandoning annual performance reviews has been greeted with accolades and some relief. Critics are optimistic that this is the beginning of the end for a costly and anxiety-inducing annual workplace ritual.
Yet some caution is required before celebrating, because the end of the annual performance review is not necessarily the end of the problems these reviews can induce.
Performance evaluations of some kind occur constantly. In heavily monitored workplaces like some call centres, workers’ performance is continually assessed by monitoring software. But even in less intrusive workplaces, your performance is being frequently observed and judged in multiple ways.
What is occurring is not the end of performance review but a change in how it occurs. Companies abandoning formal annual reviews have suggested various alternatives, from project-based reviews, to using simpler performance indicators, to the ominous sounding “continuous appraisal”.
But how confident can we be that new performance appraisals will improve on the old?
The veneer of objectivity
Many criticisms of annual performance reviews could equally apply to almost any performance evaluation process.
One point performance review critics like Christophe Dejours make is that much of what we do at work cannot be properly measured. Even the most astute managers will miss hours we spend in the office, and the innovations, relationships and the compromises we make to keep our organisations going. Other parts of work are almost permanently hidden from public view. Consider the odd experience of getting our best work ideas in the shower, or waking in the early hours of dawn in a cold sweat over a work dilemma. These experiences reveal how much our time, thoughts and emotions continue to work outside the office. What gets measured is only a small part of what we put into our jobs.
When it comes to judging performance, performance appraisal research provides a long list of actual and potential biases. These include cognitive biases like the “halo effect”, where a manager’s judgement about one of a person’s characteristics distorts perception of all the others; or the flattening “Veblen effect”, named after a distinguished professor who allegedly gave all his students a C. They also include entrenched social biases, such as women being stereotyped as less able to exercise authority. Given the extensive catalogue of biases, it is remarkable that annual formal performance reviews have survived for so long.
Of course, many judgements at work are subjective. But performance appraisal systems can make these judgements appear objective. Consequently, part of workers’ anxiety over performance measurement can be traced to how performance appraisals can give politicised or partial judgements a veneer of legitimacy in their organisations.
Another recurrent problem is evaluating individual performance in team-based workplaces. Employers invariably value team players, yet typically evaluate and reward individual performance. One intellectual root of individual performance management is an idea popularised by orthodox economics, which is that all workers make a distinctive contribution to production – their “marginal productivity” – that can and should be individually rewarded. But it is far from clear that this is possible in modern workplaces.
Finally, measurement can distort incentives. The flipside to the adage that “what gets measured gets done” is that important tasks that are not measured are often neglected. For example, there is increasing concern that relentless measurement of how many articles academics publish in prestigious journals undermines other crucial university activities, like quality teaching and conducting lower status research that nonetheless matters to local communities.
Where to now?
These problems are not solved by ignoring the performance question. Most people care about their work, and value being recognised when they do well.
But the problems with performance appraisal show that replacing annual reviews with more regular or flexible systems does not necessarily get to the heart of the issue, which is how to accurately and fairly evaluate what people contribute.
It is worth seriously considering whether we need systems of performance review at all. We all invariably judge other people’s performance; equally invariably, our judgements are often flawed. When do performance review processes reduce our inevitable biases, and when are they simply a costly and fraught method of reproducing them?
If organisations do retain formal performance measurement, there are no easy answers to what these systems should look like. But given the biases of performance measurement, and their emotional as well as financial costs, a good start would be to acknowledge they are partial and fallible. And to genuinely invite discussion and reflection, rather than issue a verdict.
Replacing annual reviews with new processes might be a step in this direction. But unless the new procedures work to address the flaws of the old, they could just as easily get in the way.