Employees are gaming their performance metrics, instead of doing what's best for the company

Question

I'd have expected that all employees are expected to do what's best for the company. However this appears to be an ideal that's not reached in practice, and people seem fixated on hitting their key performance indicators (KPIs) as opposed to doing what's best for the company. I've seen situations like these happen:

1. Alice wants to move from branch A to branch B. The reason is that branch B has a much larger + skilled personnel for the work she does. Branch A opposes the move because with Alice gone their output will be worse.
2. A client approaches Bob with a job. Bob insists on handling it himself, because although the specifics of the job makes other colleagues better at handling it, this specific task is still measured in his KPIs.

Logically in the first situation, branch A should not care that their output will be worse, only that the company's output improves (which it probably will). Similarly in the second situation, Bob should not care that his KPIs will suffer, but rather that the company benefits if Bob passes the job on, since the client gets a better product.

How can a manager stop his or her employees / branches from competing against each other at the expense of the company? Can this problem be tackled at the HR level ("hire people who don't compete with each other")?

Answer 1

This observation (which I have seen myself) has been one of the most unexpected thing I've experienced in my professional life, as opposed to university or school. This is a such a complicated issue, and I'd say there are very few large companies which don't have this problem.

KPIs: First, you need to start with the premises that it isn't a question of whether KPIs are flawed, but how much they are. A Classic example is a KPI around number of lines of code written. Then the programmers start writing way too much code, the quality goes down, and this impacts the company negatively. So then, you introduce rules and try and block loop holes. I tell you that it will never work. You will have developers, for instance, choose to do the tasks which will mean they write the most code. They will be forever biased towards making decisions based on beating this KPI. They will avoid jobs which are crucial to the business, because it doesn't involve writing much code and doesn't help their KPI. You will always be reacting to this behaviour. As you plug one hole, a new one will appear.

A manager might say that we need to introduce rules and regulations to stop this, but if the incentive is high enough, then you will even witness people break not only company rules but even the law. I know of electricity sales people who would turn off the electricity of home owners and tell them they had to sign up with them to have their electricity back.

An optimistic person may be expecting people to do the right thing, even if means hurting their KPI. But if there is no reward, or the incentive for unscrupulous behaviour is high enough, you can be certain a human being will do it at least some of the time.

Although I don't think you could ever fully eliminate this problem, some recommendations:

• Make a component of the salary based on the company performance as a whole
• that will help promote actions which are better for the company as a whole.
• Make salary reviews not only about KPIs. Good managers will be able to recognise how employees have contributed to the business in ways which aren't always reflected in their KPIs.

Most importantly, employees need to care about and appreciate their company, but this goes both ways. If the company doesn't care about the employees, the employees won't care about the company.

Edit Getting some terrific comments so just want to update. I am not saying all KPIs should be thrown out, but I can criticise them especially as the main component of a salary review. @anaximander has a great quote from Goodhart's Law:

when a metric becomes a target, it stops being a useful metric

@DanK goes on to expand on why this is the case in that when humans are involved once they find out about their KPI they will change their behaviour in ways which were not intended. This gives explanation to the OP.

Undoubtedly, there are KPIs which are very important and successful in motivating employees, such as in sales, but even then you will get people who start bending the rules for these KPIs.

I wanted to quickly touch on salary. All I advocate is for the wider practice of what already exists in many companies with bonuses, stock options or other perks which incentivise employees to act in the best interest of the company.

Elon Musk, if you are reading this, don't give away your companies! As those have mentioned, it's not the role of the employee in all this to have high exposure to the company's risk. But a Christmas bonus here and there, nobody will die. I've read that Elon has given at least some shares of SpaceX away to employees who went above and beyond which I think is awesome! Such an approach also promotes loyalty to the company. I guarantee you those employees will think twice about neglecting the company for their own KPIs or jumping-ship/rocket to Blue Origin.

Answer 2

This is pure and simple a management problem.

It's one of the tasks of management to set goals in such a way that by achieving their goals and getting praise and recognition, employees will also benefit the company.

If management fails at that, then of course all the employees will do what is best for them (they will do that anyway), even if it is detrimental to the company.

So the goal settng from the top is what needs fixing. There are plenty of books and training courses about that subject if you look for them.

PS. Don't fall in the trap that goals must be measurable and verifiable. Goals must first of all be beneficial to the company. If you have the choice between trying to measure things that are beneficial to the company (although hard to measure) and measuring things that are easy to measure (but of no value or even negative value) then you must measure what's beneficial to the company first. Even if it's more difficult.

Read a story quite recently where after a complete computer breakdown IT asked management what systems should have highest priority to get running again. Management says "Management information systems obviously". IT says: "Let me make this clear, your first priority is a system that will tell you that nobody is doing any work because their systems are still down? Wouldn't it be better if some people could start doing useful work again, even if you don't know what they are doing? "

Answer 3

People should not compete? Then (wow I really have to say that?) don't force them to compete!

Rishi Goel says it, hire the right at the top. Someone who doesn't

1. introduce indicators that give employees no other chance but to survive on their own, regardless of the rest
2. whine when they do exactly this.

Answer 4

I'd have expected that all employees are expected to do what's best for the company.

No offense intended, but that is extremely naive. It ignores the entire history of labour relations, well, everywhere. There are literally thousands of examples and millions of workers involved in widespread organized worker actions designed to punish the companies they worked for. Do a search for "Lordstown Syndrome" for many examples.

You should expect that people do the behaviours that are incentivized and avoid behaviours that are disincentivized.

people seem fixated on hitting their KPIs as opposed to doing what's best for the company

Since they are punished for doing the latter and rewarded for the former, it's not clear why you expected otherwise. Incentives matter.

I am reminded of a question on another site -- posted by a professor! -- about why so many college students are fixated on their grades, rather than their mastery of the material. Because they've been told their entire lives that all that matters to their future ability to live comfortably is their grades. Why would we expect them to behave any differently?

We built these Skinner boxes; we should not be surprised when the human animals in the boxes try to maximize their rewards and minimize their punishments.

I've seen situations like these happen

Oh good heavens I was at MSFT for 16 years. I could give you war stories all day. X reports to Y, who reports to Z. X is doing poorly because X lacks skill B needed by Y but has skill C, not needed by Y. Q -- who does not report up to Z -- needs someone with skill C and not B, so Y moves X to report to Q, opening up headcount for Y, who hires rock star with skill B. Everything is great, right? Everyone is doing the best for the company. Z tells Y, "what on earth are you doing? We need a goat (low performer) on our team so that we can shaft one of them at review time, as required by HR. Give the rock star a bad review; they're the least politically powerful."

Similarly in the second situation, Bob should not care that his KPIs will suffer, but rather that the company benefits if Bob passes the job on, since the client gets a better product.

Again no offense intended but be realistic. Would you do that if you were Bob? I wouldn't! The supposition that Bob will take one for the team and be happy about it is not realistic. Bob will quit and angrily badmouth his former company on social media.

How can a manager stop his or her employees / branches from competing against each other at the expense of the company?

Give huge bonuses for cooperation that leads to good outcomes, and fire people who sabotage their coworkers. Huge bonuses. Like absolutely enormous. You know, like the bonuses that CEOs routinely grant themselves at the expense of the company.

Can this problem be tackled at the HR level ("hire people who don't compete with each other")?

One day shortly before I left MSFT they gave all of us personality tests, because middle management was obsessed with faddish metrics. Those personality tests showed that, SURPRISE, the people who were hired to build code analysis tools in a highly competitive environment were (1) analytical, and (2) competitive.

Being competitive is great. People who enjoy competing should be hired for competitive industries! The trick is to incentivize competing with your competitors, not each other.

A question you did not ask but seems germane:

Should companies incentivize actions that improve the bottom line?

Yes, but to a point. A case that comes to mind was of a rock-star-quality friend who worked for Twitter who was given the task of improve user satisfaction amongst users who post many photographs to Twitter, and made good inroads on this problem. And then at review time, they were reviewed on the metric how much additional advertising dollars were spent at Twitter as a result of this work? Since that was not the task they were given, it seems unfair to review them based on this metric; said person is no longer at Twitter and is adding value elsewhere.

The real incentive system should be:

• Management comes up with a strategy
• Workers implement the strategy
• Management is incentivized to come up with successful strategies that improve the bottom line.
• Workers are incentivized to implement the received strategy.

In my example of my friend, they should have been reviewed on whether or not the user satisfaction needle moved. Management should be reviewed on whether or not that strategy led to the bottom line return on investment.

Answer 5

This is to my mind a fundamental problem with KPIs being set for staff. Whenever you give someone a particular target with a bonus for meeting and/or penalties for missing then they will naturally game the system to maximise their own performance against the measurement system in place. The art would be to design the bonus structure to direct this effect to the company's benefit but the unfortunate fact is that as soon as you start to reward based upon some metric you disrupt the equilibrium of the system that is your company in often unforeseeable ways.

As an example take all of the financial misselling scandals that there have been over the last few years. Payment protection insurance that has been sold to those that were contractually unable to claim on the policy because of their circumstances. The agent selling the policy had no incentive to make sure that the policy was a good fit. They were driven by the commission and performance rewards that just counted the number of policies sold. Once this was identified the banks and other financial institutions have had to pay out billions to affected customers. I have not heard of the employees bonuses being reclaimed.

Another performance metric that has had unfortunate side effects is the 4 hour target time to be treated in A&E in hospital. This target caused patients that were not seriously ill to be prioritised over more recent arrivals (who may have a greater need) so that they would be out of the department before the 4 hour time had elapsed.

Answer 6

Note: this answers the quoted question, specifically. It does not pretend to solve all problems of any company ever. It used to include something about KPIs, but that aspect was rather off-topic and removed later (hence the many comments related to KPIs...).

How to stop intra-company competition?

This works pretty deftly:

• Introduce a matrix structure where the vertical ("line") management only is interested in keeping their employees happy, well educated, and working, and the horizontal management (i.e., customer account managers or whatever you call them) buy man power through some virtual internal currency.
• Remove any barriers occuring when sharing employees between different parts of the company. This happens automatically if the vertical line management consists only of Cost Centers, and the horizontal management has only Profit Centers, with no mixed entities. This means a line manager would not care at all whom his people work for, as long as they work, and as long as they are happy (unhappy employees quit, which is bad for everybody involved).
• If you want to take it as far as possible, then you can split your whole company in half; the one part consisting of people doing the work, the other part consisting of people fetching new projects. Both can obviously even more divided; i.e. the worker org unit can specialize by methodology or technology etc.; the sales org unit can specialize by domain or by large customer bases.
• A project manager who needs work done has to buy the work force from the employees. This needs to be a simple process of course, you do not want to complicate things.
• A customer manager who needs tech persons to evaluate a customer request, or create a proposal, similarly buys worker time from line managers, so from the perspective of the line manager, again, it does not matter that there is no actual cash flow.

So, management has these interests:

• Line manager: wants all his people to be as productive as possible; i.e., "sell" them to project managers and avoid downtime. Also wants to make them happy. Does not care about money. The virtual internal charge rate makes sure that he is reimbursed. Needs to make sure that whatever mechanism determines said rate per person leads to a higher rate than the salary; high enough to cover all costs involved, of course.
• Project manager: cares about money, a lot. Wants enough workforce for his project, but not so much that the incoming real money cannot pay the internal (virtual) charge rate. He, as it should be, carries the brunt of the load of planning time vs. money vs. quality. Since the ICRs of comparable employees are virtual (not based on their actual salary), he wouldn't care whom he takes on board, as long as that person can do the job and is not outside of his acceptable ICR bracket.

Problems occur and are resolved between line managers and project managers: PMs have the responsibility to care first about the success of their project (at the cost of workers). LMs have the responsibility to care first about the well-being of their workers (at the cost of projects). These guys need to work out how to resolve these problems, and will do so. The workers themselves do not need to bother themselves with that quagmire; they escalate problems to their LMs and are protected by them.

Ideally, obviously.

Answer 7

Many will disagree and take offense to this answer, but this is what is really wrong with most companies. They set goals, KPI, bonuses or whatever measurement and awards you want and they are not directly linked to making a profit for the company (Assuming making a financial profit is important and people want to get paid.).

Some say it is impossible. Well it is very hard to do, so your managers better spend most of their day on it. I guess that's why in the U.S. most companies don't last 5 years. How can you run a company if you can't identify what everyone is suppose to be working on to make money? What is the point of any incentives, job description or evaluation if you don't know what people should be doing? This is what every manager's job involves whether they know it or not.

Even a decision as simple as letting someone change branches can get complicated based on the ripple-effects on the company. You could lose a good employee by not letting her switch. You could lose a good manager if the company allows all her people to leave (Maybe the manager is not so great if she can't figure-out a way to keep good employees.). Will this person's productivity change as a result of going to branch B? What if by being surrounded by like-minded people, this person's talents are leveraged and the company makes more money? Seems like a situation where Team A should get some KPI points for doing the right thing and let her go.

The problem with the "Bob" situation is putting too much emphasis on individual productivity. Don't give Bob all the credit for doing a task by himself that could be better served by an entire team (Managers should be aware of this.). Give more incentives to the team and more of Bob's KPI contingent on team performance. With this setup, Bob would be a fool to sacrifice team performance and needs to go. Let Bob be selfish and take care of Bob, but design the rules so it benefits the company. That is called management.

Leadership fails when there is any reason one department hinders the performance of another department or team. Sales people who make excessive demands on the production of the goods and services to the point it is cost prohibitive, get punished. If the company has to hire engineers to put all the bells and whistles on the product in order for you to sell it, you take a hit on your commission. It works the other way when dead lines are not met as well. I think a sales person should get a bigger slice of the pie if she has to give the bad news to the client about a deliverable missed by another department.

The easiest trick in the book is to have someone stop doing a task and if nothing happens to the bottom line, you know it was never needed. Long and short term benefits have to be considered as well. Please, don't throw in some edge-case about doing something illegal or dangerous. That goes without saying. If you don't think it is about the money, then don't cash your check. To me, it's a great way to keep score in business.

Answer 8

You don't want people who are not competitive - that generally translates into people who are not driven to succeed or excel. Some people may not compete with others, but with themselves - in this discussion I think the upshot is the same - you need people who are driven to succeed and very often that means the firm is not their primary focus.

What you want are people who are competitive, but understand their roles in terms of the firm. It is a delicate balance.

Can this problem be tackled at the HR level ("hire people who don't compete with each other")?

Not a good idea IMO.

We'll assume you've previously a evaluated a few candidates in terms of competence and they came out even, and then you let HR make the final decision, based on who is 'less competitive'.

Having HR make the decision could be disastrous: They will want to the hire "Mr. Nice Guy", ignoring other factors. HR looks at everything from an HR perspective, and they generally love "Mr. Nice Guy", all the more so if you tell them to hire someone who doesn't compete with others.

Regardless, I would strongly recommend against relinquishing to HR any power you and other people with expert knowledge have regarding hiring - it will come back to bite you. HR evaluates people based on their HR criteria - they are not experts in the actual work and its requirements - they have a different agenda than yours, and like anyone else, if you give them power, whether or not they rightfully deserve it, it will be difficult to get it back when necessary.

---

The two examples:

The case of Alice is different than the case of Bob.

In Bob's case, the solution would appear to be quite simple: If Bob is just another employee in the firm, a client should not be approaching him privately to do a job. The client should be turning to the Project Manager or whoever it is who is in charge of getting new business and/or delegating work to various staff members.

By the same token, Bob should not be accepting work on his own. He should have told the client to contact those in charge of such matters. Bob appears to have acted inappropriately by engaging himself for the job.

So in the case of Bob, management simply needs to step in and take control of the situation and give the work to the employee most appropriate. If Bob has a problem with that, Bob has a problem. (Or, you have a problem with Bob...)

If there is no management that is in charge of such matters, the fix still remains simple: Set up the appropriate management structure.

---

Alice is a different story: Alice wants to better herself by moving, and her move may well benefit Branch B. Branch A doesn't like it because it will hurt them? How is that relevant? Why should Alice be held back because her team doesn't want her to leave?

The question here is what's better for the firm at large: Will Alice add more value as part of Branch A or Branch B? Is Alice good enough/important enough that her desire to move should be accommodated, regardless of what Team A thinks? Should the company "bend over backwards" to accommodate Alice?

Again, that's a decision for management to make.

---

Bottom Line:

I don't think you've identified the issue correctly: The problem is not competitive employees - competitive employees are generally good employees, provided they know that they have to toe under when management makes a decision that might not suit them in a particular case.

The real problem is that management is not dealing with such cases appropriately. (Or there is not management in place to do so.)

Answer 9

First thing's first: people do not come to work to fulfil their selfless passion for shareholder value, they do it because co-operating on others' goals helps to achieve their own. You will never succeed in getting people to care only about what's best for the company, but you can try to change the environment they work in to get them working together more harmoniously.

Those personal goals will include things like recognition, a sense of achievement, social status or doing a good job, but by creating KPIs you're risking people feeling that the best way to get those things is to do well on their KPIs. You're creating a big signpost saying 'this is what it means to do your job well, and this is what others will notice about you'. You should think carefully about whether they have a place and how they will be used by people to get what the want.

If you want people to do things which harm their KPIs but help the business you're going to have to ask how they're going to get things they want by doing that. Are they going to be recognized positively for it? Will their social status be harmed or improved? Can they point to a specific positive outcome and say 'I did that :)'? Do they know that it will help their advancement prospects, not hinder them?

A classic article related to this, Management by Whose Objectives, is worth reading.

If you're tempted to try to change the people to fix this, remember the Fundamental Attribution Error - people tend to put too much weight on someone's innate characteristics as an explanation for their behaviour, rather than their circumstances. This happens everywhere, and the chances are that you don't have special unluckiness in who you recruit.

Answer 10

This is a very common and typical failure of the "Management by Objectives" approach.

In short, you cannot set specific goals and link them with personal advantages and then expect people to not follow them. That is pure insanity.

There are only three solutions:

One, refine the objective system, i.e. the KPIs and how they are measured. Note that subtle details of balancing will almost certainly be lost to employees - the exact percentage of what something contributes to a bonus is rarely checked, the fact that it is a goal at all dominates. This is doubly so if during the interviews, points are highlighted beyond their contribution. For example, if an employee is reprimanded for simply ignoring a goal that contributes 5% in favor of another one that contributes 50%, he will learn that the percentages are not really meaningful.

Many people believe this is the right approach. I personally think it is bonkers.

Two, base all your KPIs only on team or company results. Basically, instead of this bonkers (personal opinion) attempt to break down company goals somehow into an interlocking grid of personal goals that will magically satisfy them all in the strategic weights you assigned... well, basically, put your actual company or organisation unit goals as the KPIs for everyone.

This is an obvious and straightforward approach that many people reject because of perceived unfairness of some people contributing more to the resulting KPIs than others. Some of these people are employees who feel like their contribution doesn't matter and thus do not experience the intended motivational effect.

Three, abandon the whole system. Bonus systems have been debunked in dozens of studies and cynics say that they are mostly upheld because the people who would decide to abandon them profit considerably from personal bonuses.

Answer 11

Management has to change.

You ask people to think about the greater good but, what is the "greater good"?

How do you translate the greater good to what a person actually does in their position in the company? You split it up in tasks and sub-tasks etc. etc. until you arrive at a package of responsibilities that you call a job. That job has KPI's.

Yes, you as management have decided that the way those people act, is the way the company wants them to act.

If those KPI's are not benefiting the company, management has done their job wrong and the system needs to change.

Answer 12

First, it isn't logical for people to act against their own best interest and instead act in the companies best interest. Even assuming that was the way people thought, it would not be in the best interest of the company for them to do so! The phrase "above my/your paygrade" isn't just CYA nonsense. You don't want your lowest paid, least educated, least experienced newest hire making critical decisions about what is and is not in the companies best interest.

But it's not the way people think. At best people have a "what's good for general motors is good for me" attidude. Given a clear indication that what is good for the company is bad for them, people will choose their own good.

So, no the problem of people putting their own team above the company and themselves above the team can't be solved by HR. That is actually looking in entirely the wrong direction.

What you can reasonably hope for is for your employees to be competent, and to use their work time to complete the tasks set to their hands in a reasonably efficient manner. HR can possibly help with that.

What HR can't help you with is determining goals and values, that is management. Management conveys their opinions on what they value most via compensation -- salary and bonuses. If management gives a bonus for tying shoelaces, expect to see a lot of employees wearing shoes with laces, which they will then tie.

It's not the employees job to determine whether tying shoelaces is best for the company, all they know is that they will get paid X for tying shoelaces and X-y for doing something else. Management wants shoelaces tied, so they will be tied.

Answer 13

How can a manager stop his or her employees / branches from competing against each other at the expense of the company? Can this problem be tackled at the HR level ("hire people who don't compete with each other")?

Reward for goal vs. activity. If moral behavior and profit is the goal then pay for that, hire people who can make money in their sleep. If you don't pay enough and have a complex set of rules and incentives it's the road to ruin: like water, everything will find it's own level, where it floods to is not where you want to be.

---

Proof/Examples:

Sometimes the reverse of logic is what gets the desired results, it's the perception of the people involved. I've explained to people that they didn't know about the work being done and if they continued to interfere with it I would make a complaint. So they came by more often with slightly better advice.

In an article by RetailWire called "Study finds rude associates sell more in luxury stores" they lead with:

"At least at luxury stores, the ruder the sales staff, the better for sales, according to new research from the University of British Columbia’s Sauder School of Business. The study found snobby associates reinforce the reputation of high-end, posh labels as privileged for the social elite.". UBC News Release.

None of the comments beneath the story support this premise, nor the credibility of the professor or the school of business involved. The concept is wholly rejected.

The Wikipedia webpage "Perverse Incentive" gives many examples, here are a few:

• In Hanoi, under French colonial rule, a program paying people a bounty for each rat tail handed in was intended to exterminate rats. Instead, it led to the farming of rats. See also the Cobra Effect.

• Bangkok police used tartan armbands as a badge of shame for minor infractions, but they were treated as collectibles by offending officers forced to wear them. Since 2007, they have been using armbands with the cute Hello Kitty cartoon character to avoid the perverse incentive.

• Providing company executives with bonuses for reporting higher earnings encouraged executives at Fannie Mae and other large corporations to inflate earnings statements artificially and make decisions targeting short-term gains at the expense of long-term profitability.

• In building the first transcontinental railroad in the 1860s, the United States Congress agreed to pay the builders per mile of track laid. As a result, Thomas C. Durant of Union Pacific Railroad lengthened a section of the route forming a bow shape unnecessarily adding miles of track.

Other examples:

• Tragedy of the commons

  The tragedy of the commons is a term used in social science to describe a situation in a shared-resource system where individual users acting independently according to their own self-interest behave contrary to the common good of all users by depleting or spoiling that resource through their collective action.

• Information asymmetry

  In contract theory and economics, information asymmetry deals with the study of decisions in transactions where one party has more or better information than the other. This asymmetry creates an imbalance of power in transactions, which can sometimes cause the transactions to go away, a kind of market failure in the worst case. Examples of this problem are adverse selection, moral hazard, and monopolies of knowledge.

• Community rating

  In a community rated market, the insurer may not calculate premium on the basis of the risk factors attaching to the particular person wishing to purchase an insurance contract, but rather the risk factors applying to all persons within the market as a whole. Thus, in a community rated market, the insurer evaluates the risk factors of market population, and not those of any one person when calculating premiums. Some form of risk equalization also often exists in a community rated system.

• KISS Principle: "Keep it simple and straightforward."

  KISS principle states that most systems work best if they are kept simple rather than made complicated; therefore simplicity should be a key goal in design and unnecessary complexity should be avoided. This can be interpreted two opposite ways: "It seems that perfection is reached not when there is nothing left to add, but when there is nothing left to take away".

Answer 14

The obvious answer is that you, as management, should give up all of your incentives and split them up among the people who are adding value to the company.
1) It will show people that you are willing to put the company ahead of your own personal gain
2) It will show the company that you're likely grading people with the wrong metrics.

(also it will show you that you're just as unwilling to do what you think your reports should be willing to do)