Shifting trends in performance management systems are giving rise to greater collaboration, motivation and employee engagement, writes David Rock
Companies have long spent considerable energy trying to develop scalable systems for keeping track of, motivating and rewarding the performance of their employees. In the 1970s, Aubrey Daniels coined the term “performance management”, and so began a universal experiment to make this process more objective and scalable. Inspired by the “management by objectives” movement, companies started to get employees to specific goals and then rate efforts on a one to five scale against these goals, annually. It seemed logical to try to be objective and give people specific feedback about their performance. The hope was that this would help pay people fairly, motivate everyone and help remove the worst performers.
However, there was a flaw in the plan. Human beings are not necessarily logical when it comes to receiving feedback from others. Our brains are built to pay close attention to the social environment. From the moment we are born and for our first dozen years or so, our survival is dependent entirely on the social world around us. We get nothing from the wild and everything from other people. As a result, we have been wired to pay close attention to changes in our social resources, including being deeply anxious about being assessed by others. Managers using a one to five rating system soon found that giving people anything except the top rating resulted in intense debate, as people experienced “social threats” to their sense of status or fairness, for example. This created a lot of unhelpful discord for otherwise very good employees. As a result, managers tended to avoid this conflict and rate their people in the middle of the curve.
In the 1980s, GE tried to solve this problem with the idea of a forced distribution curve. Unfortunately, telling managers they had to place people at different points on a one to five scale exacerbated the problem of employees feeling assessed, only this time it was worse. A recent study in one large organisation showed that 50 per cent of the employees were “surprised” by their rating, and of those, 87 per cent were negatively surprised. These people displayed a 23 per cent drop in engagement, and they were the better employees – for example, those expecting an A but given a B. Many of these folks had been getting A’s their whole life – you can imagine the arguments people had with their boss.
“We have been wired to pay close attention to changes in our social resources, including being deeply anxious about being assessed by others”
Part of the challenge with a forced curve was that it also felt like the game was rigged. At school when you are ranked, in theory, everyone could get an A based on how hard they worked. Not so in the workplace. Now you had people competing for ratings, which reduced collaboration. While collaboration in the 1980s might not have been so critical, today it is essential for getting things done.
What about the other reasons companies installed these systems? Did this approach pay people fairly? One study from research firm I4CP showed that 75 per cent of people thought these systems were unfair. Another study from CEB showed that two-thirds of the highest paid people were not that firm’s top performers. Did these systems motivate people? Well, companies who removed them saw engagement – a good proxy for motivation – go up not down when the system went away. Did the systems help remove the worst performers? Perhaps this is an area where the approach had its uses at times, such as when making wide-scale layoff decisions.
For decades, companies have tinkered with this basic system, trying to change the labels, reduce to four ratings or change the bonus systems, but it’s only in the last decade or so we’ve seen any dramatic evolution of performance management. According to research firm Bersin by Deloitte, 70 per cent of companies are considering significant changes to their performance management systems in 2015. We’ve been tracking this trend for five years and are seeing two significant trends emerge. I will outline one and focus more deeply on the other.
One intriguing performance management innovation involves keeping the one to five rating system but “changing what is valued”. Companies such as Johnson & Johnson and Pepsi have applied a more broad approach to performance, valuing the “how” as well as the “what”. A number of other firms are experimenting with this idea. They might have a one to five rating for goal achievement, and a one to five rating for how they achieved the goals, or for how well they aligned to the culture, values or other specific behaviours. This seems to be a significant improvement and addresses rewarding behaviours, not just outcomes. At the same time it creates a lot of work, even more work than current single-rating systems, and still has the challenge of employees having a number to react to.
With the complexity of this approach it is hard to execute more than once or, at most, twice a year. Annual goals might have worked 20 years ago, but with so much change today it is hard for goals to be relevant for more than a quarter. An annual event seems to inhibit everyday feedback, sending a signal to managers to discuss performance at more regular intervals. Even quarterly feedback today does little for younger generations craving to learn something useful every week from their boss. And any discussion about performance with one manager is strange when that manager hasn’t been involved in the half-dozen emergent teams an employee has contributed to.
In response to these shifting performance management trends, in the last five years we have seen significant acceleration around removing the idea of ratings altogether, and instead focusing on getting managers to have regular conversations about the employee’s performance and development.
In 2010, fewer than five large companies in the US had made this move. By the end of April 2015, at least 43 large companies in North America, many of them global firms, and representing well over a million employees, have removed formal ratings from performance management. This includes Juniper Networks, Gap, Microsoft, Adobe Networks, Cargill, Dell, Deloitte, PWC and many other household names. These firms no longer give people a fixed rating or place people on a performance curve. Instead, there is an emphasis on regular, quality interactions between managers and their teams, focused more on growth and development than assessment. From a research perspective, focusing people on trying to improve rather than prove themselves will activate more of a “growth mindset”. A number of studies show that priming people to have a growth mindset (as opposed to a fixed mindset) has a significant impact on, amongst other things, how well people set goals, process feedback and learn from others. So not only does removing ratings reduce the threat response people feel, it also focuses people more on growing and learning – a key to adapting to today’s environment.
“While collaboration in the 1980s might not have been so critical, today it is essential for getting things done”
Perhaps this idea might seem like a fashion for today. Let’s look at the data so far. Of those who have transitioned away from ratings, we found just one company who took steps in this direction with a pilot then changed their mind. And we found a second company who, after going through a merger, decided to put ratings back in, to help them deal with too many people perhaps. The forced ranking system does have an advantage – if you have a glut of talent, and your focus is on separating people out versus motivating and growing everyone, a forced ranking system might have its use for a period of time.
Aside from these examples, every other firm who has made the transition is continuing on this path. Additionally, every month, more companies are announcing this change. The idea of removing ratings appears to be at an acceleration point. We suspect that the trend will continue as more companies are able to share a year of results with a new performance management system. Consider the trends driving this change: these include the need to be more agile, increase collaboration, be better at growing talent, and more effective at motivating the people you have. When Microsoft removed ratings their collaboration scores skyrocketed. None of these trends are likely to go away or slow down any time soon. We strongly suspect this is a movement, not a fashion.
It’s all about the quality of conversations
The big reason that companies are making this change is they want their managers to talk to employees more about their development and generally grow them everyday, not once or twice a year. Millennials crave development, a sense of meaning and career growth. Standard performance systems encourage managers to only talk to their people about growth once or twice a year. Of the companies we have surveyed so far, 100 per cent of them say removing ratings has increased how often managers are speaking to their teams. At the same time, we believe more work should be happening to define what companies mean by “quality conversations”, giving managers clear frameworks and tools for the conversations they are expected to have, not just generic training.
“Focusing people on trying to improve rather than prove themselves will activate more of a ‘growth mindset’”
As of April 2015, we have studied around 20 companies closely who have transitioned away from ratings. It is interesting how much variation there is in their strategy. Some firms have a lot more structure, for example, defining what they want managers to talk about, how often, and in what way. Other firms are letting managers decide how often they check in on performance and not providing anything more than general guidance. No one is getting rid of pay for performance or differentiating pay; they are mostly going back to manager discretion to solve that problem, often with some guidance. Under this model, so far there tends to be the same or better differentiation of pay. Managers are letting people know where they stand by talking about their performance against their goals and other important variables, just without simplifying all of that to one number.
While there are many different strategies, what is consistent is that companies who have made this transition tend to be very anxious about it beforehand and enthusiastic about it afterwards. Their excitement is palpable – they believe they are saving millions of dollars, cutting out tens of thousands of wasted hours, and making employees happier. Measurement is showing that employees working without formal ratings are more engaged, more collaborative and are talking to their managers more. Many companies who have made the change are out on the speaker circuit sharing what they have learned about performance management. They recognise they are still learning lots about how to execute this new approach, but overall they believe they are on an important path. And so a new experiment begins.
Top 5 steps for HR executives
- Build a business case thoughtfully. It’s not just about asking your people, “Do you want ratings or not?”. Look at all the true costs of current systems and weigh up the benefits carefully. For example, there are invisible costs in reduced engagement of employees who are negatively surprised by their ratings.
- Study what performance management strategies other companies have tried, but know you need to forge your own path. Every firm has quirks that require a slightly different approach to performance management.
- Build a new brand for performance management – give it a whole new name. People need to know this is a whole new approach, not just a few tweaks.
- Develop a clear statement about the mindset you want people to have. Define how you want managers and employees to think and act differently.
- Expect to take 12 months to build your plan. It can be done a little faster with more effort, but a year is a realistic timeline to launch a whole new approach.