Throwing another person at a broken process doesn't fix it. It just gives the problem a new babysitter.
When friction shows up on a team, the reflex for most owners is to hire. More hands feels like progress, and a rising headcount looks like growth on paper. But if the real problem is that nobody knows exactly what they own, a new hire walks straight into the same mess. Here's what the research says about unclear roles, what Amazon's restructuring actually signals, and a five-step test to run before you post your next job.
What the research says about unclear roles
Role ambiguity measurably drags down performance
A landmark meta-analysis of decades of workplace research found a -.21 correlation between how ambiguous an employee's role is and how well that person performs in it (Tubre and Collins, 2000). In plain terms, the less clearly a job is defined, the more consistently performance suffers, across industries and job types.
Hiring doesn't fix a correlation like that. The new person inherits the same undefined role the last person had, and the math says they'll underperform in it too, no matter how talented they are.
Less than half of employees know what's expected of them
Gallup's 2025 workplace engagement data found only 47% of employees strongly agree they know what's expected of them at work. That number has been sliding even as headcount at most companies keeps climbing.
If under half your team can answer "what is my job?" with a confident yes, adding headcount doesn't move that number. It just adds one more person to the confused half.
A bad-fit hire costs far more than the salary
SHRM estimates it costs roughly six to nine months of an employee's salary to replace them. Separately, the Work Institute's 2025 retention report found that 75% of departures were preventable.
When a hire doesn't work out (and undefined roles are one of the most common reasons why), you're not back where you started. You're down six to nine months of salary and back at the beginning of the search.
New hires can't fix a role nobody designed
I once worked with a business owner who had a habit. Every time friction showed up on his team, he threw labor at it. A process broke down, so he hired someone. Two people kept colliding on the same task, so he hired a third to referee between them. Revenue was healthy, so on paper every one of these moves looked like growth.
What actually happened is the friction never left. It just got more people standing inside it.
The real problem was architecture, not capacity
Nobody on that team had a role with a clear edge to it:
- No clear decision rights
- No clear ownership of an outcome
- No clear sense of where one person's job ended and someone else's began
Every new hire walked into the same undefined space the last one had. Within a few months, this owner was spending his week exactly the way he had before he hired anyone: babysitting disputes, refereeing who owned what, and re-explaining decisions that should have already been settled.
The fix: redesign the roles you already have
My suggestion was simple. Before you hire again, redesign the roles you already have.
- Give each role clear outcome ownership, so a person is responsible for a result, not a vague list of tasks.
- Build a real feedback loop, so people get direct, unambiguous signal about whether they're doing the job right instead of guessing.
That's not a hiring problem. It's an architecture problem, and hiring your way through an architecture problem just means paying more people to work inside the same mess.
This is the mistake that's easiest to miss, because it doesn't look like a mistake. Headcount going up looks like growth. But a role nobody ever designed doesn't improve with more people crammed into it. It just gets more expensive to run, and slower.
The verdict: what Amazon's manager cuts really say about ownership
The decision: In early 2025, Amazon CEO Andy Jassy told senior leaders to increase the ratio of individual contributors to managers by at least 15% by the end of Q1 2025. His stated reason: too many management layers had produced what he called "pre-meetings for the pre-meetings for the decision meetings," a structure where it had become genuinely unclear who owned a decision at all.
My verdict: Jassy was solving the same problem as the owner in the story above, just at Amazon's scale. He didn't say the fix was hiring more managers to manage the confusion. He said the fix was fewer layers and clearer ownership.
Why most people read it wrong: The headlines treated this as a layoffs story about cutting costs by removing management jobs. But Jassy's own language points somewhere else. He said the goal was to "clarify and invigorate their sense of ownership" and "drive decision-making closer to the front lines." That's a role-design story wearing a layoffs headline.
Your version: You may not have management layers to flatten. But the underlying question is identical to the one Jassy asked at well over a million employees: does everyone on this team know exactly what they own, and exactly who decides when there's a disagreement? If the honest answer is no, that question comes before your next job posting, not after it.
How to tell if you need a new hire or a role redesign: a 5-step test
Run this before you write another job description. It takes less time than a single round of interviews.
- Name the friction precisely. Is it a capacity problem (not enough hours in the day) or a clarity problem (nobody's sure who owns what)? Most owners assume capacity. It's usually clarity.
- Map decision rights before you map job duties. For every recurring decision on your team, whose call is it, and does everyone actually agree? If two people would give you two different names, that's your real problem, and a new hire won't settle it.
- Give every existing role one measurable outcome, not a list of tasks. "Handles customer support" is a task list. "Owns resolution time under 24 hours" is an outcome someone can actually be held to.
- Build the feedback loop before you build the headcount. People do a job better and faster when they get direct, regular signal on whether they're doing it right. That loop costs nothing to build, and most teams have never had one.
- Only hire once the redesigned role survives two weeks without a body in it. If the friction disappears once ownership and decision rights are clear, and the only thing missing is capacity, that's a real hire. If the friction is still there, hiring will just relocate it to a new person.
This week's move
Pick the one role on your team causing the most friction right now. Before you post a job for it, write down who owns the outcome and who makes the call when there's a disagreement. If you can't answer both questions clearly, that's the fix. It isn't the hire.
Related reading: if your team keeps colliding over who said what, start with Your Team Isn't the Problem. Your Communication Is. If the hiring decision itself keeps getting pushed off, read The Interest You're Paying on Every Decision You Won't Make. And before you add payroll, check Why Can't You Pay Yourself This Month?
Order The Brown Box
The Brown Box: How to Fix What's Broken in Your Business Before It Breaks You is out now, and Realign Staff and Roles is chapter seven of the TURNAROUND method for exactly this reason. Hiring feels like progress because headcount is a number you can point to. Role clarity doesn't show up on a spreadsheet the same way, so it's the fix most owners skip, right up until it's the reason they can't stop hiring. That chapter walks through how to fix the role before you fix the roster. Order now on Amazon or learn more about the book.
Already read it? A two-line review on Amazon does more for the book than almost anything else.
Not sure whether you need a hire or a redesign? Get in touch.