In the first year of my engineering degree, I thought that going to the pub was an acceptable substitute for studying differential equations. Consequently, when I received my last supervision report before exams, it read “1.0 / 5.0 – Linda will most likely fail.”
As students, we are used to seeing our performance quantified on exams and report cards. While exams are easy to grade, as soon as more abstract and subjective factors like “Effort” and “Behaviour” are brought into the picture, quantifying performance becomes a tricky matter. In the workplace, we see many such abstract factors at play when assessing employee performance and engagement.
In this tutorial, I will describe the history of quantified feedback and current trends such as employee analytics.
Quantifying employee performance is no new feat. Frederick Taylor, the inspiration behind Henry Ford’s legendary management practices, pioneered Scientific Management practices. Scientific Management analysed the performance of industrial labourers with the intention of increasing productivity.
Time and motion studies involved production managers intensely monitoring workers with a stopwatch, documenting each discrete step they took, and the time taken to do this. From this, the most efficient procedure for completing a task was determined, and the time taken to do this calculated from the sum of individual steps. Based on this benchmark, managers would calculate expected production capacities that each worker was expected to attain.
Cartoon satirising the dehumanising effects of Taylorism.
Taylor often cited the fable of a pig iron worker called Schmidt who was able to carry 47 tons of pig iron a day rather than the usual weight of 12.5 tons, and was rewarded with pay of $1.85 a day instead of the usual $1.15.
In hindsight, we see that Schmidt was being exploited, doing nearly four times the normal amount of work for less than twice the normal pay. We can see how Taylor’s methods helped to improve efficiency by reducing the unit cost of labour. However, Scientific Management made many workers’ lives miserable by ignoring the human aspects of work. Quantified by their level of production, people were essentially treated as machines.
In modern knowledge work, employees are no longer treated as machines because they’re precisely valuable for their qualities that cannot be automated. Those qualities include creativity, flexibility and curiosity.
The field of human resource management has evolved to promote good treatment of workers to allow them to perform at their best. However, measuring performance is still key to people management, and this usually involves some sort of quantification of intangible traits. Let’s look at some reasons why managers would quantify employee performance:
Recognising good performance/ addressing areas for improvement
Tracking goal attainment
Comparing performance between employees
Linking performance with bonus pay
Matching development with job requirements / company values
Visualising team strengths
Determining training needs
There are many benefits to managers introducing quantitative performance metrics, and there are various case studies of how this has been implemented in practice.
One of the most notorious performance management practices is ‘stack ranking’, where managers are forced to rank their employees on a bell curve. GE adopted stack ranking under Jack Welch, with the name “Rank and Yank”, where the top 20% of employees were rewarded and the bottom 10% fired.
This is a particularly ruthless example of quantifying performance, and there are plenty of softer options. Google allows employees to take an active role in performance reviews. They set their own goals and define a series of quantifiable results that will allow them to reach them, and are held accountable to these by their managers. Managers and peers provide additional reviews and rate the employee on a 5 point scale, incorporating some elements of the 360 review.
Stacked or forced ranking has been used at GE, Microsoft and Yahoo
As is expected, the matter of attempting to quantify performance is extremely sensitive, especially when it is linked to external factors such as bonus pay and recognition schemes. Employees may feel that they are being reduced to a set of numbers, which can cause them to feel objectified. Negative reviews of employees may feel like a personal attack.
Studies have actually shown that reviews often reflect more on the traits of the manager than the employee being reviewed. Managers should therefore ensure that the metrics they are using address employee performance rather than personality – employees should be able to act on feedback.
Questions that ask what you would do rather than what you think are shown to be particularly effective. Deloitte recently introduced a set of statements and asked managers how much they agreed with these on a scale of 1-5. An example is, “Given what I know of this person’s performance, I would always want him or her on my team”. This is called the Likert scale, and measures a positive or negative response to a statement. It is the same technique that we use at Peakon to measure employee engagement.
As the trend of big data sweeps through every imaginable domain, it only makes sense that its principles are appearing in human resource management. If we take the 3 V’s definition of big data – volume, variety and velocity, we can see that companies are starting to bring these concepts to feedback practices.
Accenture and Deloitte have recently realised that annual performance reviews are ineffective because they are not frequent enough – managers tend to focus on the most recent developments. They have replaced them with more frequent and short surveys to gather more data and provide better performance tracking.
These two may have grabbed a lot of attention as first movers (for companies of their size and stature) in ditching annual reviews. However, as I hope I have shown, we expect this approach to become ubiquitous as businesses become more adapt at applying data to HR and new tools help facilitate this process.