Performance Management (PM) is the most common way employees’ contributions are measured and valued. Since the bad old days of “rank and yank” PM has been refined repeatedly, but rating performance against key indicators or goals remains at the heart of modern people management.
Unfortunately, despite all attempts at reform, Performance Management just doesn’t work.
Many leading companies are now dumping it entirely, but most still cling on, persuaded by the common sense assumptions that make it seem essential. Here we explain why what is commonly believed is actually non-sense, and why you should stop “managing performance”.
A brief survey of the history of modern people management reveals a deep vein of pessimism about human nature.
It starts with Nineteenth Century industrialists using clocks and whistles to regulate working hours and rest breaks. “Enlightened” employers at this time were building model villages, schools and hospitals, while using the twin threats of unemployment and eviction to impose rigid codes of morality on their workers.
The dawn of the Twentieth Century saw the evolution of “scientific management” practices (such as Time and Motion studies, often called Taylorism), the Human Relations Movement (pioneered by the ex-patriot Australian Elton Mayo and others at the Harvard Business School), and, from the 1940s onwards, the use of Performance Appraisals.
All these approaches to managing people proceed from the assumption that left to their own devices workers are inclined to be inefficient, and to attempt (perhaps quite rationally) to get paid as much as possible for doing as little as they can get away with.
Papering the Cracks
From the 1960s Performance Appraisal began to evolve into Performance Management (PM), largely because appraisal on its own was found to be ineffective. Older systems used crude carrot and stick methods (awarding bonuses to the highly-rated and firing “under-performers”), arbitrary standards and opaque, even secret evaluation methods liable to lots of subjective variation as well as abuse. However, rather than abandon the whole approach, managers with a largely pessimistic view about workers simply tried to improve it.
Perhaps the most infamous managerial pessimist of them all is Jack Welch, celebrated “super-CEO” of GE from 1981 to 2001. He pioneered the use of the “vitality curve”, otherwise known as stacked ranking, and the “rank and yank” approach to employee incentives. Stacked ranking is a policy that obliges managers not only to rate employees against KPIs, but to distribute ratings on a curve, so that each year 10–20% of employees will receive a top rating (and thus all or most of any bonuses on offer), 70–80% will be given a merely adequate rating, and 10% get a low rating. As a result, employees are rated relative to each other, not to a set standard. Rank and yank goes a step further, requiring managers to fire the lowest 10% every year.
At the same time Welch was launching his ranking and firing system, he was also determined to create an innovation culture by introducing a flatter, less bureaucratic management system, less formal interpersonal relationships, and a “small business experience at a large corporation”.
The Welch phenomenon, with all its apparent contradictions, shows how the competing ideas of Taylorism (with its emphasis on efficiency and managerial control) and Human Relations (emphasising what we would now call leadership, workplace culture and employee engagement) became reconciled in the managerialism promoted by MBA curricula.
Modern Performance Management reflects this synthesis. On the one hand, it employs Human Relations thinking in its focus on the performance conversation and setting individual goals, allowing for professional growth and rectification of performance issues during the evaluation cycle. On the other hand, PM also uses performance appraisals, and sets goals related to productivity. Typical performance management plans include output-based goals (KPIs) combined with increasingly-sophisticated measures of things Mayo would approve of, such as 360-degree assessments of team work, leadership and other “soft skills”. The clearest indication that many managers remain pessimists is the continuing widespread acceptance of the “common sense” belief in the need to provide incentives in the form of bonuses and promotions, and to link these to ratings against individual outputs or outcomes.
In the 35 years since Jack Welch inaugurated the era of the all-powerful CEO, human resources leaders have worked hard to refine the methods he pioneered. A growing number of law suits and business failures led to a critical re-examination of older forms of performance appraisal and performance management, but despite superficial changes in purpose or method, contemporary people systems are largely the descendants of Taylor, Mayo and Welch.
The lack of real innovation in people management stems from a set of shared assumptions that seem like common sense, but are just biases supported by an ideology.
The power of these assumptions is enhanced by the way they dovetail with the same ones underpinning modern mainstream economic theory (often called neoliberalism or economic rationalism).
Some of the most important assumptions include:
- Rationalism — the belief that behaviour, especially economic behaviour, is guided by rational assessments of self-interest. Ergo if you pay someone more they’ll work harder, but they will never do more than they have to do.
- Money makes the world go round — since money is a “universal medium of exchange” every good can be exchanged for it, and nobody will contribute effort without it.
- Principle of Least Effort — People will always do the least amount they need to in order to get what they want. Since employees will rationally want to get the maximum financial return for the least possible effort, they have to be bribed, threatened, cajoled and manipulated to maximise productivity.
- Everything that matters can and should be measured, and measurement is essential to productivity. Conversely, if something cannot be measured it doesn’t matter.
- Competition drives innovation, improvement and quality.
While explicit stacked ranking and rank and yank approaches have largely gone the way of the dodo, the pessimistic assumptions underlying them endure in practices like “up or out” (favoured by large professional services firms), performance management through KPIs and other output-focussed goals, the widespread use of performance-based pay, and the continuing obsession with numerical ratings.
Unfortunately for pundits, managers and workers alike these assumptions are broadly false, and the systems built on them (in which companies worldwide currently invest upwards of USD5 billion a year) produce mostly perverse results.
The mounting evidence of how badly PM backfires has led some very prominent businesses to abandon PM entirely in recent times, including Accenture, Microsoft, Adobe, Deloitte and, most surprisingly, GE. Many of these companies are abandoning annual performance appraisals, and in GE’s case even ratings.
The evidence against PM, performance appraisal and the whole edifice of neoliberal managerialism, both anecdotal and empirical, is piling up.
Anecdotally, leaders have been increasingly swayed by a series of high profile business disasters.
Microsoft, the colossus that bestrode the information revolution almost from the start, went into sharp decline at the beginning of the new millennium. For at least a decade it fell farther and farther behind the innovation bleeding edge, came to be regarded by consumers as the very essence of monolithic corporate power, and lost market share along with share value at an alarming rate. In a 2012 Vanity Fair article, Kurt Eichenwald stated that every single serving and ex-Microsoft employee he interviewed, without fail, described their version of stacked ranking as the principal reason behind the company’s almost-inconceivable fall from grace.
In the USA, Ford, Goodyear, Capital One, Microsoft and Yahoo (among others) have been subject to high-profile civil law suits arising from their use of stacked ranking methods. Worse, these suits often point to the way rating systems can easily facilitate the operation of biases relating to age, gender, parental status or other markers of diversity.
Recent empirical research is now uncovering the reasons why performance management systems tend to strangle innovation, waste money, and worsen the effects of bias.
The biggest impact on employee engagement and thus performance relates to the way being rated triggers fight-or-flight responses in the human brain. Even when employees receive positive ratings, the experience itself is sufficient to create defensive behaviours. These include managing risk by avoiding stretch goals, ensuring goals are generic or vague and thus easily met, and minimising competition. Microsoft employees interviewed by Eichenwald described how high performers would try to avoid working alongside other high performers, to avoid be down-ranked. As a result, it became difficult to assemble star teams for high-value projects. Because of this fight-or-flight response, competition between employees tends to dampen innovation and improvement, despite the common assumption to the contrary.
The other negative impact of employee appraisal relates to the complexity of human motivation. Despite the many core assumptions human resource managers take for granted, people are not solely, or even centrally motivated by monetary reward. Humans have complex drives, including needs for reward with an emotional aspect (such as recognition), a desire for fairness, safety needs (that drive defensive behaviours), bonding or social needs, and intellectual needs (to understand and explore).
Focussing on performance rating undermines these drives, especially when the assignment of ratings is experienced as arbitrary or unclear.
Even economics, the discipline that informed much people management thinking over the last several decades, is starting to move away from its core assumptions. Instead of relying on models of economic behaviour that assume people are “rational maximisers”, emerging sub-disciplines such as behavioural economics are starting to look at how complex, often emotional rather than rational drivers, shape economic behaviour.
Given the weight of anecdotal and empirical evidence it is time for people management to become far more holistic in its treatment of people, and to give up on mechanistic approaches such as performance ratings.
Rather than keep tinkering with a performance management system in the hope of fixing its inherent problems, the solution is to recognise that the assumptions behind PM are false. Ditch them completely and start over with new, optimistic assumptions supported by the evidence.
An optimist will work from the following starting points:
- Employees have complex needs, and these can be addressed
- Recognition and reward can take many forms, and they don’t all cost money
- People are mostly willing to work hard, provided their effort has a valuable purpose
- Not everything that matters needs to be measured, provided you measure those things that can tell us if we’re heading in the right direction
- The best source of innovation and improvement is collaboration
This alternative outlook is not new, it has simply been overlooked by mainstream managerialist thinkers. Many of the core ideas of a collaborative approach to people management can be traced to the work of Dr W. Edwards Deming, an American engineer, statistician and management consultant, most famous for his contribution to Japan’s economic recovery after WWII and popularising of the Plan-Do-Study-Act (or PDSA) approach to production management. Deming diagnosed “seven deadly diseases” of ineffective business leadership, including “mobility of management” (a direct critique of the MBA-era idea that managing is a discrete skill independent of knowledge of a particular industry) and reliance on employee evaluation and annual performance reviews.
Deming argued that focussing on individual performance is a mistake because only 15% of problems in a company arise from employee errors. 85% of problems are caused by leadership shortcomings and systemic problems in the company’s oversight, production and quality assurance processes. So using performance evaluation to drive productivity and profitability is simply blame shifting. We can further infer that focussing on individual performance leads businesses to ignore systems thinking. It also promotes disengagement (because PM violates the need for fairness when employees carry responsibility for systemic problems), and defensive practice.
Deming’s insights, developed since the 1940s, were published in 1982. However, the dominance of powerful CEOs like Jack Welch, and the prominence of Taylor and Mayo in management education have obscured Deming’s work. He is now vindicated by the failure of the ranking approach in people management and the growing empirical evidence.
Models for an alternative people management approach can also be found in successful non-English-speaking economies such as Germany, where collaborative methods (such as co-determination) and continuous education to update skills are normal and supported by government policy.
A collaborative approach that replaces PM in human resources management will include:
- Shifting focus from performance to professional development
- Setting goals to do with personal improvement and contribution to shared objectives
- Replacing performance ratings with assessments of growth in one’s role and contribution across the business
- Empowerment of direct managers and employees to set development goals, and to identify and help rectify systemic issues
We will discuss all of these features of a collaborative people system in forthcoming articles.
You can also read this on Medium