Very interesting article. But I wonder, can these four questions really capture the performance of employees? It seems better than many systems but perhaps it is too narrowly focused.
I've worked with Deloitte employees on a few occasions, in my experience their managers (and often several layers of managers) have worked directly with each member of staff and know them well. The annual reviews seem to an outsider to be an important formality for justifying what is already relatively clear.
I wonder, will four questions will grow into six, eight, twelve whatever, which will then require a form or series of forms to be completed & some sort of redundant moderation to ensure they aren't dealing with individual perceptions again etc etc, eventually evolving into what they have just done away with.
I've seen this happen at companies before. The other side I've seen is:
HR bod: sorry, you can't give everyone a 5 on your team, someone has to have a 1
Manager: but, I've hired my team sensibly!
HR bod: sorry, you can't give everyone a 5 on your team, someone has to have a 1
Yep currently got that.
HR: You may only put up one person for promotion per year
Me: But I have 5 people in the team. Its unfair.
HR: Sorry. Only one.
At least it's being up for promotions, not having to finger one for remediation/termination. The binning scheme where every team needs a super star and a low performer (a la Jack Welch) is atrocious and the smaller the unit it's applied to, the more likely that the curve fits.
I've seen (in Banks) this applied like a religion and it leads to situation where the "low performing" member of Team A would be solidly in the middle of the pack of Team B, if not the "high-performing" member of that team.
If you're exceedingly lucky in that kind of environment, there's enlightened management running interference and manually balancing team to avoid having to can a good person because their team is overall good.
I worked in healthcare briefly and saw a lot of the same. We'd have a handful of people assigned to new teams and magically 3-6 months later half of them would be part of a round of layoffs. It got to the point where anyone who got reassigned during non-obvious times (excluding a business unit failing or a small acquisition or something) almost immediately started looking for another job and they usually quit within a month anyway.
Yes, seen that. One variation I experienced was separating retention rank (how critical is this person to the team, dept., etc.) from performance rank (how effective and skilled is this person). That was a bit of a painful exercise as it was very clear to me that the organization was sifting for a reduction in force despite high performance after having culled the low performers.
It makes sense on the most basic level of "we have 5 high-performing individuals but we only have enough work to support 4."
If you have a group of people who all get 5's on a review but you are required to let one of them go, how do you do it in a way that isn't picking a name out of a hat, or the first person to get stuck in traffic and be late to a meeting?
But one knows what the numbers can be used for when assigning performance review numbers to people. So instead of picking a name out of a hat, the process is to pick a number out of a hat corresponding to a name?
1. Given what I know of this person’s performance, and if it were my money, I would award this person the highest possible compensation increase and bonus.
2. Given what I know of this person’s performance, I would always want him or her on my team.
3. This person is at risk for low performance.
4. This person is ready for promotion today.
Questions 2-4 seem OK, but question 1 is problematic. I've been conducting data analysis on questionnaire data and people can't be expected to imagine "if it were my money". It's not their money, it won't be taken from their paycheck. Basically the stakes are not real. Which in turn will create in upwards bias.
Yes, but what does the question actually capture? What's the incentive not to agree with that statement? Or maybe everyone puts "4" instead of 5 to hedge their bets?
Don't want to deride the question without seeing someone from Deloitte expand on that, but the "as if it were own money" seems redundant at best.
I don't think there's an incentive in reviewing others, apart from the general "greater good". The "own money" part is probably to get people to think a bit more honestly, rather in a "sure, why not" way. I agree that it doesn't sound very effective.
I'd be glad to give everyone a raise with someone else's money -- when there's no downside to doing so. I think the point is to remind you that there's a cost associated with that.
It is perfectly fine to applaud Deloitte, sadly, this concept won't spread.
HR departments are so entrenched with annual review bullshit that even if they call it a different name and put a new "spin" on it, it will just end up being even more disappointing shit-work for everyone.
I understand the sentiment, but the concept is spreading. Adobe, Netflix, Juniper, Accenture all have or are re-thinking their performance management models. Atlassian has been influential here too, and one of the architects of their revamp has now moved to the Broad Institute. CEB, which is very influential in the HR world, has adopted these new approaches and are actively recommending them to clients. Samuel Culbert, a UCLA professor, has been very visible on this front, and people like Josh Bersin and Marcus Buckingham (co-author of linked HBR article) are really well-known in HR circles.
It will, of course, take a long time to undo the Jack Welch and command-oriented mindsets. However, I do think there's hope that strategic HR departments can re-orient companies toward processes & policies that care less about compliance with HR bullshit and more about actual coaching, feedback, and development.
(Disclaimer: I re-designed an employer's performance framework & process, so I am biased toward folks who are trying to push HR toward people-oriented, trust-based programs that are designed for the vast, vast majority of employees who don't need old-school performance "management.")
I do performance management for Fortune 500 company. Literally two weeks ago we began redesigning the model.
It may sound that it would be difficult to change what has been going for years, but there are several factors that make it easier than it seems so:
- Normally performance management process is managed by few people even in companies sized in 10,000's of employees. Even if ultimately the decision is made by HR executives, it is easier to make such decision when there are so few directly involved people. These people are normally part of corporate, and they just provide the process for local HR that usually has no real say in how performance management is done.
- Both corporate HR and local HR are tired of this too. We have been hearing these points for long time. Every single year enforced/recommended (amongst other points) distribution brings heated discussions yet nothing is done. Executives hear this.
- Talent Management system vendors are becoming more flexible and allowing different models for performance management. Such system is needed, but no one would ever sign up for developing it themselves.
In overall, I think it is a matter of time till this spreads. Promoting the wrong people or blocking the good ones is very expensive.
> In overall, I think it is a matter of time till this spreads. Promoting the wrong people or blocking the good ones is very expensive.
I have worked for several companies that follow Trend-driven Management and would not be surprised to see this spread amongst the CEO's and execs of major companies.
Just be 'radically transparent' (a la Bridgewater) and don't be hesitant to give feedback in REAL-TIME. That's the only way to consistently improve. It's surprising to me that places like Deloitte haven't embraced this yet...instead favoring traditional political correctness and 'professionalism' and frustratingly clunky methods like this.
Your comment is a perfect example of why companies don't adopt this.
Critical feedback is too often construed as 'unprofessional' or 'inconsiderate' when it's simply a subjective statement on the way someone does something.
If we as a people could take feedback more constructively rather than misconstruing it all as some kind of pejorative statement of our character, we wouldn't have this issue.
So really the problem is that we don't have ENOUGH professionalism. Too much personalism and not enough professionalism. The personalism gets in the way of true, transparent feedback.
I have a verbal 'tick': I ask 'Why?' all the time.
When I ask it, I mean: "What was your thinking that lead to you making that statement or claim?"
I learned some time ago that regardless of MY intent in asking it, somehow, people act as if they are being attacked or disagreed with.
I'm just looking for context and trying to understand the reasoning process.
If we, as a culture, can't even ask "Why?", then I have to assume that (professional) real-time subjective assessment would be a big non-starter for people. Culturally, in the US, anyway.
A large part of my motivation to start my own company is to define my own culture. How is one to progress without asking 'why'? How is asking 'why' once a year (and in abstract, politically-correct terms) supposed to help us? Why not do it better, faster?
And for the commenter above, how is asking 'why' not professional? Clearly there are professional ways of going about it.
I wonder why so many people think like that. Especially when it's understood that when bad things happen in business, it's 'just business as usual' (e.g., getting fired is often said to be 'just business as usual'...nothing personal). Then why can't we be more transparent about feedback in the workplace?
For the record, I think what you're calling a 'tick' is a virtue.
> For the record, I think what you're calling a 'tick' is a virtue.
Very kind of you to say. The automatic way I point the question at anybody who crosses my path may say otherwise.
I have noted that in personal argumentation, in the US and Japan, that the word is not often not used as a tool for inquiry. I've seen it wielded more as an accusation or bludgeon, and it may be this dual-use that make people sensitive to the word itself.
The child's response to any answer with a "why?", is truly a really excellent algorithm for getting to the bottom of things.
That is weird. They ask to rate not the person, but your feelings against the person.
Yes, this very much simplifies manager's job - but remembering Kahneman's Thinking Fast and Slow, people are notoriously skewed at measuring things through their feelings. A lot of personal factors, including levels of sugar in reviewer's blood, heavily affect this type of 'judgment'.
I can see how Deloitte may be happy about spending less time on bullshit, and I applaud to that.
I can also see how this new system is no different in consistency of the results compared to the previsions one - both are pure bullshit, so weighted random number generators work approximately the same when asked for 4 of 10 weighted random numbers.
However, I really do not see this as progress. This is more of a regress - they see the problem, but they offer the
solution that makes the managers slightly happier, and does nothing to precision or fairness of the system.
What we need to do it stop blindly applying a sales-oriented performance rating system to engineers.
We need specific set of quantified criteria for each group, and resulting rating should be calculated independently of the rater.
We do not need more manager-oriented performance systems. We need to look at these systems form the point of the employee who is being rated.
This is exactly what I've being saying for years of that project basis individual competition, it's a a popularity contest, albeit at high school level, defines today's most IT organization.
The four questions seem sensible but having employees' evaluation judged solely by their team leader seems guaranteed to create a culture of managing upwards and groupthink.
Ex-deloitte employee here. I left before this change in performance review took place, but here's some thoughts & context based on my experience:
- People may be concerned that relying on four questions may be too narrow, but at the end of the day, regardless of which questions are officially on the performance eval form, those are the questions that matter. At Deloitte, I had to consistently find new projects every 2-6 months, and the question of "Given this person's performance, I would always want him or her on my team" is not hypothetical. Project manager and partners answer this question implicitly all the time when they hand-pick their team members, now it's just answered more explicitly. At the end of the day, if you're an employee, what's more important to your career: that your boss thinks you're an Excel whiz or that she enjoyed working with you and will put you on an interesting, challenging project next month?
- I think some of the criticism that these questions essentially about your likability or ability to "manage upwards" are fair. But for a firm, whose business is all about providing client service, evaluating people on their people's skills for "managing upwards" seems like a reasonable proxy for their ability to "manage clients", which really is a valuable skill to the company. On the other hand, for engineering orgs, these four questions may need to be rephrased - (e.g. off the top of my head: "do I want this person responsible for delivering and maintaining critical production services").
I especially agree with your second point. While many performance measurement programs do screen for the most likable or personable people, this is actually intended in consulting. Consulting is about selling work, and you will sell more work if people like you and you're pleasant to work with. And you're right, managing upwards can be rephrased as "getting people to do what you want them to do without any authority to make them do it" -- basically the entire job of a management consultant.
That said, many consulting companies have overly-complex performance measurement systems. At the end of the day, if you're not very good, you won't get staffed on projects and you will be laid off. If you are good, push for a promotion and you'll probably get it. None of that changes based on the measurement system you use; because people at either end of the spectrum will/won't game whatever measurement system you enable. And IMO if you can get the same results with a lot less overhead, it's a win/win.
TLDR; In 2014 HR conducts a survey to find out what the people they hired think of HR's other surveys and find plenty of work for HR to spend the next year on and publishing about to arrive at a four question survey for release in 2015.
44 comments
[ 4.2 ms ] story [ 107 ms ] threadhttps://hbr.org/2015/04/reinventing-performance-management
Very interesting article. But I wonder, can these four questions really capture the performance of employees? It seems better than many systems but perhaps it is too narrowly focused.
But they can quite possibly capture them well enough to assign bonuses and trigger discussions about promotions.
Use any search engine 'deloitte scandal'
I've seen (in Banks) this applied like a religion and it leads to situation where the "low performing" member of Team A would be solidly in the middle of the pack of Team B, if not the "high-performing" member of that team.
If you're exceedingly lucky in that kind of environment, there's enlightened management running interference and manually balancing team to avoid having to can a good person because their team is overall good.
If you have a group of people who all get 5's on a review but you are required to let one of them go, how do you do it in a way that isn't picking a name out of a hat, or the first person to get stuck in traffic and be late to a meeting?
2. Given what I know of this person’s performance, I would always want him or her on my team.
3. This person is at risk for low performance.
4. This person is ready for promotion today.
Questions 2-4 seem OK, but question 1 is problematic. I've been conducting data analysis on questionnaire data and people can't be expected to imagine "if it were my money". It's not their money, it won't be taken from their paycheck. Basically the stakes are not real. Which in turn will create in upwards bias.
Don't want to deride the question without seeing someone from Deloitte expand on that, but the "as if it were own money" seems redundant at best.
HR departments are so entrenched with annual review bullshit that even if they call it a different name and put a new "spin" on it, it will just end up being even more disappointing shit-work for everyone.
There is always good consulting work for connected people out there.
It will, of course, take a long time to undo the Jack Welch and command-oriented mindsets. However, I do think there's hope that strategic HR departments can re-orient companies toward processes & policies that care less about compliance with HR bullshit and more about actual coaching, feedback, and development.
(Disclaimer: I re-designed an employer's performance framework & process, so I am biased toward folks who are trying to push HR toward people-oriented, trust-based programs that are designed for the vast, vast majority of employees who don't need old-school performance "management.")
It may sound that it would be difficult to change what has been going for years, but there are several factors that make it easier than it seems so:
- Normally performance management process is managed by few people even in companies sized in 10,000's of employees. Even if ultimately the decision is made by HR executives, it is easier to make such decision when there are so few directly involved people. These people are normally part of corporate, and they just provide the process for local HR that usually has no real say in how performance management is done.
- Both corporate HR and local HR are tired of this too. We have been hearing these points for long time. Every single year enforced/recommended (amongst other points) distribution brings heated discussions yet nothing is done. Executives hear this.
- Talent Management system vendors are becoming more flexible and allowing different models for performance management. Such system is needed, but no one would ever sign up for developing it themselves.
In overall, I think it is a matter of time till this spreads. Promoting the wrong people or blocking the good ones is very expensive.
I have worked for several companies that follow Trend-driven Management and would not be surprised to see this spread amongst the CEO's and execs of major companies.
Critical feedback is too often construed as 'unprofessional' or 'inconsiderate' when it's simply a subjective statement on the way someone does something.
If we as a people could take feedback more constructively rather than misconstruing it all as some kind of pejorative statement of our character, we wouldn't have this issue.
So really the problem is that we don't have ENOUGH professionalism. Too much personalism and not enough professionalism. The personalism gets in the way of true, transparent feedback.
When I ask it, I mean: "What was your thinking that lead to you making that statement or claim?"
I learned some time ago that regardless of MY intent in asking it, somehow, people act as if they are being attacked or disagreed with.
I'm just looking for context and trying to understand the reasoning process.
If we, as a culture, can't even ask "Why?", then I have to assume that (professional) real-time subjective assessment would be a big non-starter for people. Culturally, in the US, anyway.
And for the commenter above, how is asking 'why' not professional? Clearly there are professional ways of going about it.
I wonder why so many people think like that. Especially when it's understood that when bad things happen in business, it's 'just business as usual' (e.g., getting fired is often said to be 'just business as usual'...nothing personal). Then why can't we be more transparent about feedback in the workplace?
For the record, I think what you're calling a 'tick' is a virtue.
Very kind of you to say. The automatic way I point the question at anybody who crosses my path may say otherwise.
I have noted that in personal argumentation, in the US and Japan, that the word is not often not used as a tool for inquiry. I've seen it wielded more as an accusation or bludgeon, and it may be this dual-use that make people sensitive to the word itself.
The child's response to any answer with a "why?", is truly a really excellent algorithm for getting to the bottom of things.
http://www.washingtonpost.com/news/on-leadership/wp/2015/07/...
Yes, this very much simplifies manager's job - but remembering Kahneman's Thinking Fast and Slow, people are notoriously skewed at measuring things through their feelings. A lot of personal factors, including levels of sugar in reviewer's blood, heavily affect this type of 'judgment'.
I can see how Deloitte may be happy about spending less time on bullshit, and I applaud to that. I can also see how this new system is no different in consistency of the results compared to the previsions one - both are pure bullshit, so weighted random number generators work approximately the same when asked for 4 of 10 weighted random numbers. However, I really do not see this as progress. This is more of a regress - they see the problem, but they offer the solution that makes the managers slightly happier, and does nothing to precision or fairness of the system.
What we need to do it stop blindly applying a sales-oriented performance rating system to engineers.
We need specific set of quantified criteria for each group, and resulting rating should be calculated independently of the rater. We do not need more manager-oriented performance systems. We need to look at these systems form the point of the employee who is being rated.
- People may be concerned that relying on four questions may be too narrow, but at the end of the day, regardless of which questions are officially on the performance eval form, those are the questions that matter. At Deloitte, I had to consistently find new projects every 2-6 months, and the question of "Given this person's performance, I would always want him or her on my team" is not hypothetical. Project manager and partners answer this question implicitly all the time when they hand-pick their team members, now it's just answered more explicitly. At the end of the day, if you're an employee, what's more important to your career: that your boss thinks you're an Excel whiz or that she enjoyed working with you and will put you on an interesting, challenging project next month?
- I think some of the criticism that these questions essentially about your likability or ability to "manage upwards" are fair. But for a firm, whose business is all about providing client service, evaluating people on their people's skills for "managing upwards" seems like a reasonable proxy for their ability to "manage clients", which really is a valuable skill to the company. On the other hand, for engineering orgs, these four questions may need to be rephrased - (e.g. off the top of my head: "do I want this person responsible for delivering and maintaining critical production services").
That said, many consulting companies have overly-complex performance measurement systems. At the end of the day, if you're not very good, you won't get staffed on projects and you will be laid off. If you are good, push for a promotion and you'll probably get it. None of that changes based on the measurement system you use; because people at either end of the spectrum will/won't game whatever measurement system you enable. And IMO if you can get the same results with a lot less overhead, it's a win/win.