Performance Management: The hype and the facts

The hype is here, and the word is that performance management is dead. Microsoft, Accenture, Deloitte, and now even GE have all scrapped the traditional performance review. And so should you. At least, that’s what the headlines say. But before we all leap on the bandwagon and follow suit, let’s pause for a second. Because all is not what it seems, and some of the headlines are just downright wrong. Performance reviews may be changing in ways that are definitely desirable and long over-due, but they are far from dead. So let’s sort the facts out from the hype and see what is really happening, and what you need to do about it.

How performance reviews have evolved

What we call traditional performance reviews are actually a relatively recent innovation, having risen to prominence in the mid-1950s. Management by Objectives or Management by Results were the buzzwords back then, and the focus was on setting precisely defined targets and then measuring progress towards these. There was great hope in this new approach and the way it enabled pay-for-performance, supposedly ushering in a new era of meritocracy in which people who worked harder and got better results would be rewarded for it.

Unfortunately, a fundamental challenge quickly became apparent. Appraisal ratings were often inaccurate or not perceived to be fair representations of people’s actual performance. This was due in part to the fact that managers – like all people - are riddled with biases that make it difficult to objectively rate others’ performance. And compounding this was the fact that it can be difficult to discern what proportion of someone’s results is due to their effort and skill, and what is due to broader factors, such as the market or the processes that individuals have to work with.

In response to these issues, the 1980s and 90s saw the rise of a range of approaches that tried to expand and enhance performance reviews. New methods such as Total Quality Management, and behavioral and competency-based appraisals broadened the targets people were evaluated against, in order to try and provide a more holistic picture of performance. Companies tried to ensure more objective ratings by using processes such as ‘calibration’, in which managers who oversaw similar groups reviewed one another's employee performance ratings. And 360-degree feedback also began to be used, again as a way of trying to create greater objectivity, this time by collecting the ratings and perceptions of lots of people.

The really big development in this period, though, was the rise of Forced Ranking systems, which involved a subtle but significant shift of focus. In addition to evaluating how well employees had delivered on their objectives, forced ranking also required managers to compare employees’ performance to that of their peers. In its strictest form that meant rank ordering people from best to worst, and in a lighter form meant ensuring that certain proportions of people were categorised as over- and under-performing. The most famous proponent of this was GE, and in the US in particular it became a common appraisal method.

Problems remain

Yet problems remained. Competency and behavior-based appraisals are still vulnerable to rating biases. Calibration tends to be extremely time-consuming and it is unclear whether it succeeds in making performance ratings more accurate. And while 360-feedback systems may deliver multiple perspectives on individuals’ performance and are often useful sources of developmental suggestions, they are generally not effective as appraisal tools, as they too are prone to rating biases.

Forced ranking, on the other hand, does succeed in eliminating the generalized over-rating of people, which is why it has proved so popular. Yet it can also bring with it a whole new set of challenges. Dick Grote, who literally wrote the book on forced ranking, has shown how it can undeniably be useful in small doses, especially in businesses that struggle to get managers to rate anyone as under-performing. But he has equally noted that prolonged use of forced ranking systems can breed employee cynicism, lowered engagement levels, and even counterproductive behaviors.

Knowing that perfectly adequate performers – people who have done everything asked of them – can be rated as underperforming in order to meet quotas for classifying people as poor performers, can undermine confidence in the fairness of the system. And there are anecdotal reports (most famously from Microsoft), that by limiting the number of people who can be top-rated, forced ranking has unintentionally incentivized behaviors such as not recruiting potential high performers who might become rivals, deliberate undermining of peers’ performance, and a general reduction in collaboration.

Moreover, hanging over all the innovations in appraisals has been the fact that they often do little – if anything - to improve performance. This is partially due to the fact that appraisal conversations typically happen only once or twice a year, and that is just not frequent enough in most cases to have a significant impact on behavior and performance. And it is partially due to the fact that time-pressured managers often lack the skills or inclination to really push the performance improvement aspect of the appraisal.

Little wonder, then, given all these issues, that a recent Bersin study found that of 2500 companies in 90 countries, only 8 percent said that the annual review process was “an effective use of their time”.

Big headlines

So we have reached the point where people are trying to go back to the drawing board and reinvent performance appraisals. And in the last 6 months, the pace of change seems to have reached a tipping point. In April 2015 Harvard Business Review led with an article showcasing how Deloitte were ‘reinventing performance management’ with a ‘simpler’ and more sophisticated system. In July, a Washington Post headline announced that, “Accenture will get rid of performance reviews and rankings”. And then in August came the big one – GE, probably the most famous proponent of forced ranking – with headlines announcing that they were ‘killing annual performance reviews”. Appraisals, it seems, are all but dead and buried.

But look beyond the headlines, and things are not quite as they may appear. Not one of the big name companies supposedly ditching appraisals are getting rid of performance reviews entirely. Most of them still require managers to rate people in some way, and many are still linking these ratings to reward. So what, you may ask, is actually changing?

What is changing

Looking beyond the headlines reveals one thing immediately: that although the unhappiness with traditional annual appraisals is clear and growing, just what they need to be replaced with is far from certain. That said, there are two developments that everyone seems to be in agreement with.

1. An end to forced ranking. Forced ranking is an over-used method, especially in the US, but there is a growing consensus that it needs to be replaced. In fact, many of the headlines you will have seen about companies abandoning appraisals have actually been about them stopping forced ranking, rather than stopping appraisal ratings altogether.
2. Replacing annual reviews with more frequent meetings. In an effort to both get more accurate appraisal ratings and drive performance improvement conversations, an increasing number of firms are replacing annual reviews with more frequent check-ins. Once a month or quarter seem to be the most common timeframes being used, with the idea that managers can give more timely feedback and performance improvement suggestions to staff.

These two developments are both very welcome and overdue, but are not solutions in their own right, because they raise further questions. Three big ones stand out:

• Having ditched forced ranking, firms are still left with the fundamental challenge of how to get accurate performance ratings; so should they abandon rating people altogether?

• If they do continue with ratings, then how can they improve the accuracy of these ratings?

• And how can they make sure that managers are using the more frequent appraisal meetings to good effect, to genuinely improve people’s performance?

Do we need performance ratings at all?

If you ask firms why they collect performance ratings you usually hear one or more of three different answers.

• We need to know who our best people are. This is the weakest argument for having performance ratings, because there are other ways that managers can indicate who their best performers are. For example, managers could simply ‘tag’ the highest performers in their team with a high-performer label. So you do not need to rate everyone in order to identify who your best people are.

• We need to let people know how they are doing. There is certainly some evidence that some employees – and often the high performing ones – are motivated by being told they are doing well. Yet you do not need performance ratings to give people a sense of how they are doing. Verbal feedback can also let people know how they are doing, as can tracking progress against performance targets. However, while there are alternatives to the traditional performance rating, there is underlying concern that these alternatives may not fuel the most competitive people in quite the same way. There is thus a lurking suspicion that by getting rid of ratings, firms may inadvertently demotivate some of their highest performers.

• We need to know who to reward. This one is trickier. Performance ratings are used in many businesses as part of a formula for driving bonuses – to ensure that those who work hardest and get the best results are rewarded for this. There are other ways to ensure this, though. One simple way is to train managers in how to motivate and reward people through the use of non-financial incentives. Another way, which firms such as Microsoft are trialing, is to give managers more autonomy in how bonuses are distributed. For example, managers can be given an overall bonus pot for their area and then left to decide how this is doled out amongst their team. Some firms are also creating a pot for on-the-spot bonuses, or micro bonuses, to give managers greater freedom to reward effort and results at the time they happen. The argument against these solutions is that managers may fail to be objective in evaluating how well people have done, and thus biased in how they dole out the bonus pots. And there is a lack of research yet on whether local bonus pots can indeed work, so the jury is still out on this one.

For each of the reasons to give performance ratings, then, there are alternatives. You do not have to use performance ratings. And this is why we have seen a growing – though still very small – number of firms publically renounce performance ratings altogether. Yet for all the publicity surrounding these moves, it is important to remember that these firms are essentially experimenting. There is just not enough evidence yet to say for certain that removing performance ratings will do good. Or at least, not do any harm. Particularly, in terms of how people are rewarded. And that is why the majority of firms have not leapt to follow suit: because it is a big move with uncertain impact.

How to improve performance ratings

If you want to keep performance ratings, then, but do not want to go down the route of forced ranking, what is left to help you assure the quality of ratings? Well, you have four main options.

• Rating training. The most common solution found in firms is training managers in how to make better ratings. And the most popular type of training involves educating managers on the unconscious biases that can impair their objectivity and reduce the accuracy of the performance ratings they make. While there is evidence that this can indeed improve the accuracy of ratings, it cannot remove biases altogether. Moreover, recent research suggests that there is a more effective type of intervention, too, called ‘frame of reference’ training. This involves asking managers to rate the performance of people in case studies, so is effectively calibration and practice rolled into one. Again, this is not a complete solution because it cannot remove biases altogether. But it is a step in the right direction.

• More frequent ratings. The first option is ask managers to rate people more frequently – the idea being that this puts a stop to the tendency for annual ratings to be biased by recent events. In theory, then, more frequent ratings should produce a more accurate evaluation of performance across the whole year. Once a quarter seem to be the most common timeframe used here, though there are reports of businesses asking managers to do this as often as once a month.

• More specific ratings. The second option is to distinguish between different types of performance and to ask managers to rate each of these separately. For example, some firms are distinguishing between bottom-line performance (results produced), leadership performance (how well people lead and develop their teams), and team performance or organizational citizenship (how well people work with peers and other stakeholders). The idea is that more specific ratings allow for a more granular picture of performance and avoid the situation where overall ratings are distorted by one particular aspect of someone’s performance.

• Just plain different ratings. The third option is to go down a different path and ask managers not to rate performance, but something different. The most high profile proponent of this approach to date has been Deloitte, who ask managers to answer four questions. First, managers are asked to rate individuals against two statements: “I would award this person the highest possible increase and bonus” and “I would always want this person on my team.” And then they are then asked to answer yes or no to two statements: “This person is at risk for low performance” and “This person is ready for promotion today.” The idea here then, is that by asking a different sort of question, you many reduce some of the ratings bias that occurs.

Just as with the companies abandoning ratings altogether, the firms trying out these new approaches are garnering a lot of headlines. But just like the companies abandoning ratings, the firms trying new ways of rating are also essentially experimenting and there is a lack of evidence that they do genuinely lead to more accurate and less biased performance ratings.

That said, there is a strong and growing feeling that using a combination of more frequent ratings and just plain different ratings, supported by rater training, may produce better and more reliable performance data. It is certainly less harmful than using forced ranking, and less time-consuming and expensive than calibration. So right now, it looks like the best option.

How to ensure performance is improved

The final question here is how to ensure that performance reviews genuinely do something to improve people’s performance. In theory the move to more frequent appraisal meetings should result in more coaching and feedback for employees. Yet the question remains how to make sure that managers do actually coach their people, and do so in a way that is effective.

The most common solution is to provide support for managers. There has thus been an upsurge in the number of businesses investing in providing training for managers in how to coach their people. This can be in the form of workshops, or online modules. In addition to this training, or sometimes instead of it, firms are also using scripts to help managers run review sessions. These scripts lay out the areas that managers need to cover in the review meetings. And to make sure the scripts are followed, some organizations are asking managers to submit data about each of the topic areas - so to answer questions such as “what has the individual done well” and “what could they do better”.

One further development here is the use of technology to enable employees to receive light-touch, continuous feedback. Vendors such as Reflektive, Impraise, Engagedly, Workday, and SuccessFactors are all embedding ongoing feedback into their performance management tools. And some large organizations are developing their own solutions. GE, for example, has recently launched an app called “PD@GE” (“performance development at GE”). The app is used to record what is discussed and committed to in appraisal discussions, and which employees can use to give or request feedback from anyone else at any point.

One slight caution here is that again, for all the hype, it is still early days and we do not know for certain that these innovations will help. After-all, what they all try to do is to increase the amount of feedback given, but feedback does not in itself guarantee development and improved performance. It is fine telling someone what they need to do better, but people often need help to change. So while the new feedback-enabling technologies may indeed prove useful, they are unlikely to be maximally effective unless part of a broader solution that includes things like coaching training.

Recommendations

What the hype and headlines have been correct about is that things are changing. What they have glossed over is that what we need to do now is not yet clear. For the moment, based on the evidence available, what we can recommend is this:

• If you have been using forced-ranking for more than a year or two, stop it.

• Replace annual appraisals with more frequent performance review conversations. Somewhere between once per month and once per quarter should do.

• Provide managers with training and scripts to follow for these regular review meetings.

• Collect some ratings in the regular review meetings – do not leave it all to year end – and ask for multiple, ‘different’ specific ratings rather than just one overall rating.

• Consider using a technology solution to make both giving and gathering feedback, and submitting performance ratings easy.

Performance management, then, is far from dead. Rather, it is getting a new lease of life, and shows signs that it might finally become what it always should have been: a useful and effective tool. Moreover, there are signs that for once the cries of ‘revolution’ may indeed be justified. For the re-imagining of performance management is leading us re-emphasize the importance of good old fashioned management skills. Never mind the leader as strategic visionary; this is all about the leader as people manager. The leader as someone who can motivate people and help them become better at what they do. And reinforcing this is the fact that the emergent technologies are not just being used to make employee appraisals easier, but to also provide managers themselves with feedback on their managerial skills.

The fundamental change occurring beneath the hype, then, is a revolution in reverse. A return to fundamental management skills. The delayering of organizations that has occurred over the past 30 years will not make this easy. Managers are busier and more stressed than ever before. But the promise and prize awaiting is more motivated and capable employees. And that, is something worth fighting for.