Picture a dealership where every department is excellent at what it is measured on.
Marketing is measured on leads, and it produces more leads every month. The BDC is measured on appointments set, and it sets more. Sales is measured on closing percentage, and it closes a higher share of what it touches. Finance is measured on gross per transaction, and the number climbs. Management is measured on units this month, and the month gets hit.
Now look at what each of those competent departments did to get there. Marketing bought cheaper, lower-intent traffic, because volume was the target. The BDC booked anyone who would agree to a time, because set was the target, and the showroom filled with people who were never going to buy. Sales learned to avoid the hard, viable customer and work the easy one, because percentage was the target. Finance added friction to every deal and burned a little trust each time, because gross was the target. Management sacrificed next quarter's customer to this month's board.
Nobody was incompetent. Nobody was dishonest. Everybody did exactly what the measure asked, and the business got worse in direct proportion to how good they were at it.
That is the sentence I keep coming back to: optimising the wrong objective with extraordinary competence produces worse outcomes than mediocrity would have. The better the people, the faster the damage.
Gaming the number is the small problem
Everybody knows what happens when you pay people on a number. They find the shortest way to the number. Pay the BDC on appointments set and every "maybe Saturday" becomes an appointment. Pay sales on closing percentage and the hard customers quietly stop getting logged as customers. The number goes up, the report looks great, and it is not telling you anything.
That is gaming the system, and every manager has seen it. It is also the easier problem. You can catch it, and when you catch it the number is fixable.
The harder problem is the dealership in the first paragraph, where nobody is gaming anything. Everyone is doing exactly what they were asked, honestly and well. The numbers are real. And the business is still getting worse, because the numbers were never the business. Each department was given a number that stood in for what the owner wanted, chosen because it was easy to count, and each department chased its number past the point where chasing it started costing the store. No one cheated. There was nothing to catch. That is what makes it dangerous.
Different objectives, the same behaviour
Give five departments five different targets and you will see them do the same things. Protect the headcount. Hold on to the good leads. Resist any change that could make the department look smaller. Treat a handoff to another department as a loss. Those behaviours are not personality. They are what any competent agent does inside any system where its own measure is what it is judged on. The enemy of my number is my enemy. You can watch it in a dealership, a hospital, a sales floor, a government agency, or anywhere else people are paid against a proxy.
So when a principal tells me a department is being difficult, I do not start with the people. I start with what the department is paid to maximise and ask whether a reasonable person, paid that way, would do anything else.
The owner's objective is usually unstated
Here is the uncomfortable part. Ask the owner what the business is for and you will get a word. Profitability. Growth. Now ask which profitability. Gross this month. Net this year. Lifetime value of the customer. What the group sells for in five years. Those are different objectives, and they produce different decisions on the same deal.
Most businesses never write the answer down. So each department fills in the blank with its own number, and the system optimises for whatever was easiest to measure. Then the owner is surprised that everybody hit plan and the business does not feel any better.
The agent is not misbehaving. It is doing exactly what it was incentivised to do. The objective was never stated carefully enough to constrain it.
The deeper question
For a long time I thought the failure was that organisations chose the wrong objectives. I now think that is the easy version.
The harder version is this: most organisations have no mechanism for noticing when hitting their objectives is actively undermining the business. There is no report that says the appointments went up and the sold went down and here is the connection. There is no meeting where the question is asked. Every department's dashboard is green, and the one measure that would be red, the business against its own target, is not on anyone's screen.
That is a design problem, not a people problem, and it is the one worth solving. Not better targets, though those help. A way for the system to recognise when its own success has become the thing that is hurting it.
What to do with this
Write down the owner's actual objective, in a sentence, and make every department's measure answer to it. Put the business's own number on the same page as the departmental numbers, so that the gap is visible every month and not once a year. Pay attention to the handoffs, because that is where one department's win becomes another's loss. And when a department is performing brilliantly and the business is not, do not congratulate the department. Ask what it is optimising, and whether that was ever what you wanted.