Performance management has suddenly become hip, possibly for
the first time in business history.
Companies like Amazon, Microsoft, Deloitte and even GE are tearing up
the rule book and radically redesigning their processes. Some changes are smart, some are brave, some
are interesting, but all are missing the point.
Why PM? Why now?
A recent survey of several thousand international business
and HR leaders found that 89% had recently changed or were about to
change their approach.
Why the rush to innovate? Because businesses have finally realized that
traditional performance management does not do what it says on the can. Instead of managing how employees perform, too
many processes are:
- Bureaucratic – forms! forms! forms!
- Time-consuming – Deloitte calculated their in-house process took 30+hrs per employee
- Demotivating – 90% of conversations are focused on what went wrong in the past, leaving employees defensive and disengaged
- Downright useless when it comes to improving performance
Everyone has known this for ages, but companies still stuck
with idiotic policies like rank & yank partly from laziness, partly because there was no obvious alternative, partly
because the impact of performance management was only vaguely measured, if at
all, so nobody saw quite how bad the situation was. Fast-forward to today where HR is at last
starting to embrace evidenced-based management, and it has suddenly become
obvious that performance management needs an overhaul – in the large survey
cited above nearly 6 out of 10 executives believed their current approach
drives neither engagement nor high performance.
So, what’s been brought in to replace the traditional
Performance Reprisal? New flagship
systems generally focus on three things:
1. Agile, future-focused feedback
Instead of cramming a year’s worth of performance into a
single annual review session, the new performance management – PM 2.0 – spreads
the discussion out over anything from quarterly to weekly or even daily
sessions, either face-to-face or via internal social media. And most of the talk is intended to be about
the future, not the past.
There are a lot of good reasons for making performance
management continuous. It avoids recency
bias, where evaluations focus unfairly on what has just happened. It provides a wealth of data to mine for
trends and correlate to specific circumstances or events (the impact of an
especially demanding project, an intensive training course). It enables problems to be spotted early, and
just-in-time support offered. And –
hopefully – it makes both manager and employee more relaxed and
business-as-usual about the process, avoiding the catch-me-if-you-can system gaming
which research by The Corporate Executive Board found explains a full 2/3 of
alleged top performance.
It also makes sense to focus on the future. Crying over spilt milk or slapping backs
while reminiscing about glory days are a waste of office hours. How can I help you do better? is a
question far more likely to provoke genuine reflection, engagement and
determination to change than the usual variations on Well
done! / What went wrong?
2. K.I.S.S.
A recent survey by the Human Capital Institute found that
78% of HR and executives believe their business processes have become too
complex. PM 2.0 tackles this challenge head on. Gone
are 56-item checklists and forms as thick as old-style telephone directories
(remember those?). Instead, think elegant
simplicity or the rather coarser acronym above which I’m not going to spell out
in a public forum.
Such simplicity is often radical. Deloitte has reduced its process to just 4
questions, two of which ask managers not what they think of the employee but
what they would do with the employee (how much they would pay & whether
they would want them on their team again).
Other companies have held back from prescribing any content and instead
trust managers and employees to work around general guidelines such as “set
expectations, offer feedback and recognize strong work”. The hope is that not only will this removal
of complexity increase commitment to the process (Wow - it’s so easy!), but that it will also free up more manager
and employee time for actually getting work done.
3. Talk about pay another time
Finally, HR has woken up to the slew of research that shows highlighting
pay in performance discussions makes most of us nervous, defensive, avaricious,
aggrieved and thoroughly disengaged.
Companies like Adobe are experimenting with beyond-financial reward
approaches – shout-outs on social media, project choice, extra responsibilities
and other ways of recognizing value. Even those who stick to a traditional salary
increase/bonus approach are decoupling the paycheck determination from talking
about performance and how to improve it.
Not that they’re sure exactly how they are going to tackle the reward
issue. As a Deloitte spokesman said, we haven’t resolved this issue yet …
So, what's not to like?
Don’t get me wrong, many of these changes are good news for
business and for employees. Measuring
performance in real online or offline time is clearly superior to the once-a-year
showdown, especially for Millennials who are used to constant feedback and attention. Focusing on the future is likewise a
no-brainer; as those ads for mutual funds always say in their footnotes, past
performance is no guarantor of future returns.
Stopping pay rows discussions from infecting the rest of work can
only be a good thing. And anything that
removes bureaucracy and cuts down on the time people need to spend filling out
forms is always a winner.
But there’s something missing from all these new approaches. Can you guess what it is? Take a look at the title of this post for a
clue.
Yup, you got it: PM 2.0 is missing out on managing
performance. Here’s what’s gone wrong:
1. Not combating bias
One of the biggest criticisms of traditional performance
management was that evaluations were often subject to rater bias – ratings varied
on the basis of who was doing the rating, rather than who was being rated. Bias increased with non-specific rating
methods such as the popular graphic rating scales that asked raters to judge
people in terms of ill-defined factors like “Leadership” on generic scales from
“Very Poor” to “Excellent”. Instead of
reflecting on and evaluating an individual’s performance (a complex and
mentally taxing process), raters answered with their general impressions, many
of which reflected their own values and self-image rather than the employee’s
actual work.
PM 2.0, by removing much of the detail of the process, takes
us back to those bad old days of rater bias.
Cognitive psychology – see the wonderful book by Nobel laureate Daniel
Kahneman, Thinking Fast and Slow – reveals
that, given the choice, human will always opt for the easier thinking challenge. When it comes to performance management, we dodge
complex questions such as evaluating an individual’s contribution in favor of
easier judgments such as Do I think this
person is great?. This would not
matter if the complex and easy questions were essentially the same, but by asking
the wrong questions we unsurprisingly often get the wrong answers. Being thought a great guy or gal has very
little correlation with actual business results.
By substituting unstructured brief conversations for a more
formal process, PM 2.0 is risking exactly the kind of bias performance
management was designed to remove. There
are good and time-efficient ways to fix this – developments in technology and
psychometrics have made the very accurate Behaviorally Anchored Rating Scales
(BARS) and Behavioral Expectation Scales (BES) easier to implement, for
example. But leaving it up to individual
managers is almost certain to derail any systematic management of performance,
and may expose the company to the risk of being sued for legally-indefensible
decisions.
2. Omitting what actually improves performance
When supply chains were reinvented back in the 1980s,
businesses saw huge returns on eliminating unnecessary activities and focusing
on the essential steps – and only on those steps. While PM 2.0 does away with a lot of
inessential paperwork and meetings, it does not succeed in focusing on what
really matters – the core drivers of high performance at work.
What all the performance management re-engineers have
forgotten is that we know – with a very high degree of certainty – what drives
success at work. While the individual details
vary from job to job, what’s generally needed is for employees to like the type
of work, have sufficient smarts to meet the thinking challenges, deploy the half-dozen
competencies that strongly differentiate the best from the rest, and enjoy
working in the organization’s culture.
Bringing this into performance management means a focus on no more than
a handful factors. If performance
management focuses on improving employees’ capability in these areas, performance
will improve. If performance management
focuses on other things, success is much more of a coin-toss.
It really isn’t hard, or time-consuming, to focus on what
matters. It isn’t as sexy as thinking up
your own metrics, or as fun as free-wheeling conversations, but it does have
the advantage of being proven accurate and proven to work in developmental
terms. Too many of the new specifics –
where they exist – are unproven and therefore high risk. Deloitte claims, for instance, that asking
managers what they would do with employees rather than what they think of them
is valid because the question has higher internal consistency. I’m sure it does, but concept consistency is
only one part of assessment validity.
What’s missing is any proven causal link to performance. One might accurately observe, as Deloitte
did, that top performers enthusiastically agree with the statement At work, I have the opportunity to do what I
do best every day, but only deep, longitudinal, outcome-focused research
can show whether the relationship between that agreement and high performance is
causal and, if so, in which direction the causality works.
If performance management was a fringe business process,
perhaps this kind of freestyle experimentation would be okay, but with people being
one of if not the major driver of competitive advantage, it looks at best
foolish and at worst reckless not to use reliable data as the basis for core
management processes.
3. Forgetting the human factor
Finally, the new performance management systems just don’t
speak to what people at work need.
Employees want help in getting success and fulfillment. For older workers, this often means career
advancement; for Millennials, career passion and coaching is more important. In nothing I have seen written about PM 2.0
has there been explicit instruction to include these issues in performance
discussions, despite the fact that 2/3 of respondents to a recent HCI survey
reported that they are currently in the process of updating their engagement
and retention strategies, and that 87% see culture and engagement as a top
challenge.
Employees also want to know where they stand. Deloitte admitted that people had liked its
old system for its predictability and obvious fairness (similar attitudes have
been found across companies and industries that use structured,
behaviorally-based and future-focused systems).
The company unsurprisingly found that employees were unsettled by the
vagueness of the new system when it came to sharing ratings or detailing
exactly what information was going to be taken into account in setting pay or
promotions. A big risk of all these new
approaches is that too much is left to be worked out during implementation, and
that employees will be left in the dark, disengaged, while that happens.
Ironing out such wrinkles in PM 2.0 is likely to fall to
managers, who may well as a result find they are spending as much time as ever
on performance management meetings. And
if they are not stuck in meetings, they are having to be trained in career
management, as Adobe found after it abolished formal performance assessments in
2012 only to find two years later that both managers and employees required
extensive training in instigating and acting on personal development plans. Without clear guidelines and structures, both
sides understandably felt under-equipped to deal effectively with complex,
high-stakes conversations.
There’ s a great story in Daniel Kahneman’s book (page 229ff
of the paperback) about his early experience evaluating
performance in the Israeli Defense Force.
When Kahneman arrived, interviewers had been used to assessing new soldiers’
potential for high performance on the basis of a general conversation covering
a range of topics. Everyone liked the
process, but follow-up research had found the performance predictions were
almost totally useless. Kahneman instituted
a new approach based on a structured evaluation of a small number of discrete
personality traits, followed by a general judgement of success potential. The interviewers hated it – they called the
process robotic, boring and insulting.
But when the results came back, it turned out that not only did the
structured evaluations pretty accurately predict the soldiers’ future
performance, the interviewers’ overall judgements became stunningly more
accurate than they had been under the old free-for-all system.
Following a structured process did not
replace human intuition, it super-charged it.
There’s a lesson here for performance management. To get a great result for employees, managers
have to work a little harder, and in non-intuitive ways, while keeping
interactions simple and agile. Too much of
PM 2.0 seems designed to lessen the burden on managers, rather than to improve employee
performance. If PM stood for Pauses for
Managers, that would be fine, but so far nobody has suggested that is a core
business improvement process.
Given the mixed bag of good progress and silly ideas that is
PM 2.0, we really should be thinking about Performance Management 3.0. In a future post, I’ll talk about what that
might look like.
