Picture the call you’re about to make.
You’re dialing the recruiter. Eighteen months ago you were on a different version of this same call about the same role. Three months from now you’ll probably be back in this seat with this phone in your hand. You replaced your last department head, things tightened for a quarter, and then the same client-quality dip started showing up in the same accounts at roughly the same point in the engagement.
You’re about to spend nine months and a meaningful piece of leadership attention on another search. You’re going to vet harder this time. Reference calls, working session, paid trial. The new hire will come in stronger. Things will tighten again. Around month four, the dip will return.
This is not a hiring problem. It is a design signal.
Repeating Outcomes Are Outputs, Not Accidents
When a result returns after the person responsible for it leaves, the result is no longer a personal failure. It has become system behavior.
A surgeon doesn’t blame the scalpel when the same kind of wound keeps reopening. She looks at the technique, the patient, the conditions of the recovery room. The scalpel was used. The scalpel is not the cause.
When the same client-delivery dip shows up under two different department heads, in the same accounts, at the same point in the engagement, the dip has stopped being something a person did. It has become something the seat produces.
The reason this is hard to see is that the surface evidence keeps pointing at the person. The dip really did happen on her watch. The escalations really were her accounts. Her predecessor really did underperform. None of that is wrong. It is just incomplete.
The question underneath the surface is different. What about this seat trains capable people to produce this result.
The seat is not neutral. It has a meeting cadence. An escalation path. A relationship to revenue targets. A set of metrics it is measured against. A founder relationship with its own pattern of praise and silence. Each of those is an input. The output is what you keep firing for.
The Seat Trains the Operator Faster Than You Train Her
Watch a new department head in her first ninety days.
She walks in with grace and political capital. She questions premises. She rebuilds the team’s stand-up. She rewrites the intake form. She schedules the kind of client check-ins her predecessor stopped doing. Things tighten. You feel the difference. So does she.
Then the seat starts teaching her.
She learns which clients escalate to you directly when she pushes back. She stops pushing back on those clients. She learns which internal escalations land on her review and which quietly disappear. She routes her attention accordingly. She learns that the quarterly target is the only number that gets discussed in the leadership meeting. She begins protecting the target the way her predecessor did. She absorbs scope. She lets QA slip in the last week of the engagement. She greenlights work she would have flagged in month two.
None of this is laziness. None of it is a failure of character. It’s intelligent adaptation to a system that rewards exactly the behavior you keep firing for.
By month six she’s making the same calls her predecessor made. By month nine the client-delivery dip is back. You’re reading her resume in your head and starting to think it was a bad hire.
The hire was fine. The seat is the same seat.
Five Inputs Do the Work, Almost Always Invisibly
When the seat keeps producing the same output, five inputs are usually shaping it. They are usually invisible because they were never designed. They accumulated.
The first is the meeting cadence around the role. What the department head reviews weekly, who reviews her, what gets surfaced versus buried.
The second is the escalation path. Which problems land in her lap, which route around her to you, and what the routing teaches her about what she actually owns.
The third is the metric architecture. What number she is measured on, and what that number quietly forces her to deprioritize.
The fourth is the revenue relationship. If the quarterly target is sacred and the delivery-quality metric is aspirational, she will learn which one is real.
The fifth is the founder relationship. What you praise in passing, what you frown at in passing, and what you never mention at all become the strongest training signals in the building.
These five don’t appear in the job description. They appear in the role within thirty days of any new hire taking it. They’re why the same result returns.
A seat that produces a repeating outcome is producing it across all five of these. The person sitting in the seat is the variable. The five inputs are the constants. Replacing the variable does not change what the constants are doing.
Stop Diagnosing the Person. Examine What the System Rewards When No One Is Watching.
The move that ends this loop is not subtle. It is the shift from blame to design.
The diagnostic question changes. You stop asking who is the right person for this seat. You start asking what is this seat currently structured to produce, and is that what we want it to produce.
That question is uncomfortable because it implicates the founder. The seat was built, formally or by accumulation, under your watch. The incentives that reward scope absorption are ones you let stand. The escalation path that teaches the department head to stop pushing back on certain clients runs through your office. The metric that quietly outranks delivery quality is one you protect in the leadership meeting.
This is not an accusation. It is the structural fact. Systems produce what they are built to produce. You are one of the builders.
The exit from the loop is design work, not hiring work. Pick one of the five inputs. Redesign it. Watch what changes in the seat over the next ninety days.
What This Means in Your Week
Before you post the next requisition, run the seat through a structural read.
- Pull the last three performance reviews for the role. Look for repeated language across different people. The repeated words are the system’s voice.
- Map the escalation path on one page. List every kind of problem the role owns, and write next to it who actually decides. If the role owns the problem but you decide, name that.
- Identify the one metric the role is measured on in the leadership meeting. Identify the second metric it should be measured on but is not. Notice which behaviors that gap rewards.
- Sit with the question for one week before opening the search. Ask one peer outside the company to read your role description and tell you what they think the seat actually rewards.
- Choose one of the five inputs (cadence, escalation, metric, revenue relationship, founder relationship) and redesign it before the next hire arrives, not after.
Where This Leads
A recurring failure mode in a seat is the cheapest diagnostic a company gets. It tells you precisely where the structure is producing what no one wants. Most operators treat it as a hiring failure and pay the cost of another search, another ramp, another nine-month cycle to the same dip. The shorter path is to read the seat as data and redesign one input before staffing it again.
A Discovery Session is thirty minutes with Chandler. The conversation is built around the specific recurring outcome you keep firing for. You bring the seat. You leave with a named diagnosis of which of the five inputs is doing the most work, and what one redesign would change the output. The call is free and there’s no follow-up unless you ask for one.
