Everybody Is Responsible. Nobody Is Accountable.

WHEN DECISION-MAKING AUTHORITY AND OPERATIONAL RESPONSIBILITY BELONG TO DIFFERENT PEOPLE.

A company has a problem.

Management holds a meeting. Departments explain what happened. Employees describe the work they completed.

Marketing did its part.

Sales did its part.

Operations did its part.

Everyone was responsible for something.

Nobody seems to own the result.

How the fuck does that happen?

One way is to divide the work into departments, give each department a target, and assume that completing all the assignments will automatically produce a functioning business.

Then, when the result disappoints, everyone brings evidence that their own assignment was completed.

The evidence may be accurate.

The business may still be losing money.

Completing the Assignment Is Not the Same as Choosing the Right Assignment.

Suppose management decides to launch a new service.

Marketing develops the campaign. Sales contacts prospective customers. Operations prepares to deliver.

Everyone completes the approved work.

Customers don't purchase.

Was that an execution failure?

Possibly.

But was the service priced sensibly? Did the intended customers want it? Did they understand the offer? Could the company compete with the alternatives? What evidence supported the decision to launch?

A well-executed campaign can deliver an offer the market doesn't want.

A capable sales team can spend its time pursuing the wrong prospects.

An efficient operation can prepare to deliver something nobody has established the business can sell profitably.

You need to examine the assignment as well as its execution.

The reverse matters too. A sound commercial idea can fail because the work was performed badly.

Management is not automatically wrong because the result was poor. Employees are not automatically right because they followed instructions.

If the instructions were ambiguous, the work departed from them, or somebody knowingly ignored a problem they were expected to raise, investigate that.

The point is to identify the failure, not choose a side before looking.

Responsibility Needs Somewhere to Go.

For this discussion, responsibility means the work or decisions assigned to you. Accountability means being answerable for those decisions, the way the work is managed, and the response to its results.

Those are working distinctions, not an argument about whose management textbook has the correct definition.

A manager is told to reduce expenses but cannot change staffing, purchasing, contracts, or priorities.

A marketing employee is told to increase qualified customers but cannot see what happens after an inquiry reaches sales.

An operations manager is told to meet a deadline while another department controls the information needed to begin.

The person has an objective.

Do they have a workable way to influence it?

Evaluating someone's performance requires understanding what that person could actually control.

A manager may be responsible for meeting a financial target without having authority over the expenses, staffing decisions, or operational conditions that determine whether the target is achievable.

That doesn't automatically excuse poor performance. Managers are expected to recognize constraints, communicate risks, coordinate with other departments, and escalate decisions beyond their authority.

But the company still needs to distinguish between an employee's execution and the decisions that management retained.

If you expect someone to produce a result, establish what they can decide, what resources they can access, and who must act when the problem exceeds their authority.

What can the person decide? What can they obtain from others? What happens when cooperation breaks down?

If the answer to all three is “they'll have to figure it out,” you haven't defined authority.

You've defined somebody else's fucking problem.

The Handoff Is Part of the Work.

Marketing generates an inquiry.

Sales is supposed to respond.

Operations is supposed to deliver.

Accounting is supposed to collect.

Every transition contains information, expectations, and a next action.

Sending the information does not establish that the next person received a usable assignment. Putting a name in a CRM field does not establish that the person knows they own the follow-up.

“I sent the email” tells me an action occurred.

It doesn't tell me the handoff worked.

In the context of transferring digital projects between teams, the U.S. General Services Administration's transition guidance calls for shared goals and acceptance criteria before the first team rolls off. That is a project example of a principle I would apply here: define what a successful transfer actually requires. [1]

For the customer inquiry, what must sales receive? Who accepts it? What happens if required information is missing? Who handles an inquiry that never reaches an available salesperson?

If sales makes a delivery promise, how does operations learn what was promised—and confirm whether it can be done?

If accounting cannot bill because the order lacks necessary information, who fixes it?

Don't make every employee personally chase every department forever. Build a clear transfer and a way to resolve exceptions.

The point of a handoff is to move the work forward.

It cannot just be the place where one department stops counting it.

Everybody Hit Their Target. The Company Lost Money.

Consider a hypothetical service campaign.

Marketing spends $8,000 and generates 100 inquiries: $80 per inquiry.

Sales closes twenty jobs at $2,000 each: $40,000 in agreed work.

The approved plan assumes $1,200 in variable delivery cost per job: $24,000 for the twenty jobs.

That would leave $8,000 after the specified delivery costs and marketing spend, before fixed overhead and other company expenses.

Now suppose sales promises turnaround times the existing operation cannot meet within those delivery costs.

Operations completes all twenty jobs, but spends another $6,000 on rush labor and expedited work. The company also issues $4,000 in refunds because some customers did not receive what they were promised.

Assume all twenty jobs are completed, all associated costs are included as stated, and the customer payments have been collected net of those refunds. We are comparing the same group of jobs, not mixing this month's expenses with next month's projected sales.

This is a hypothetical analysis based on an understanding of how systems work. It is not a client result, an industry benchmark, or proof that marketing caused all twenty sales.

The remainder is not company net profit. Contribution analysis separates the amount remaining after variable costs from the fixed costs and other expenses still to be covered. Here, the campaign's specified marketing spend is also deducted explicitly. [2]

Now imagine the targets were 100 inquiries for marketing, twenty signed jobs for sales, and twenty completed jobs for operations.

Every target was met.

The company is down $2,000 on the specified costs before paying the rest of its overhead.

The difference from the planned remainder is $10,000.

What caused it?

Don't decide that from the table alone.

Who authorized the turnaround promise? Did sales ignore a known limit? Did management approve the exception? Did operations commit to a schedule it couldn't meet? Did an unexpected problem change the cost after a reasonable commitment?

The arithmetic shows the consequence.

You still have to trace the decisions.

Your Metrics Can Reward the Wrong Result.

In that example, a salesperson measured only on signed work can appear successful while creating a delivery obligation that destroys the expected contribution.

A delivery team measured only on completion can appear successful while using more resources than the sale supports.

A marketing team measured only on inquiry count can appear successful without knowing whether those inquiries fit the business.

These are possible incentive conflicts, not claims that everyone measured this way will behave badly.

People do not have to be dishonest for the measurement system to encourage poor choices.

A departmental performance measure can encourage decisions that look successful inside the department while producing an unfavorable result for the business.

That doesn't make inquiry counts, bookings, or completion rates useless.

They tell you something about part of the process.

They do not, by themselves, tell you whether the whole process worked.

Keep the useful operating measures. Connect them to quality, cost, fulfillment, and the resulting economics.

If the business celebrates a booking without examining the promise attached to it, don't act shocked when somebody optimizes for bookings.

Escalation Is Not Copying the CEO on an Email.

Suppose operations identifies the impossible turnaround promise before the work starts.

It raises the issue.

Now what?

Someone has to decide whether to change the promise, change the price, approve extra resources, delay another job, or decline the commitment.

If the issue sits unanswered while the deadline approaches, the company has identified the problem without resolving it.

An escalation needs a decision-maker with authority, enough information to make the decision, and a response while the decision can still matter.

When management overrides a recommendation, that choice belongs in the explanation of the result.

It does not vanish because operations subsequently performed the work.

At the same time, an employee cannot knowingly let a preventable problem grow and then rely on “not my department” as a complete defense. If raising the risk was part of the assignment, whether they raised it clearly and promptly matters.

Trace both sides.

What did the employee do with the information they had?

What did the authorized decision-maker do when the issue reached them?

An unanswered escalation is not evidence that the person who raised it gained authority to solve it.

Accountability Is Not Finding Someone to Punish.

A serious operating failure can occur without lazy or dishonest employees.

Two departments may reasonably believe the other owns the next step. An approval may arrive after the useful window has closed. A standard process may fail to account for a predictable exception.

If you stop at the last person who touched the work, you may never examine the condition that made the mistake likely.

Google's published site-reliability guidance describes blameless postmortems that examine contributing conditions and produce preventive actions. That is an engineering practice, not a universal employment rule. The useful principle is to understand and correct the failure rather than make the investigation an exercise in personal blame. [3]

That does not mean performance standards disappear.

A person may lack the required skill. They may ignore a clear instruction. They may conceal information or fail to perform an assignment they had the resources to complete.

Those possibilities deserve a factual review too.

The distinction is between investigating what happened and starting with the employee you intend to blame.

If the process was broken, fix the process.

If the performance was poor, address the performance.

If both were involved, pretending only one existed won't help the next customer.

Somebody Has to Own the Business Result.

For an initiative that crosses departments, I want a named owner of the combined objective.

Not somebody who personally performs every task.

Somebody who can see the result across departments, coordinate the work, resolve trade-offs within their authority, and get decisions beyond that authority to the appropriate executive.

That person needs enough visibility to understand what the business is trying to accomplish, what resources have been committed, and whether the combined work is producing the intended result.

Your business may call this person a project manager, department head, operations director, or executive sponsor.

The title is not the point.

The point is that somebody has to be responsible for coordinating the complete outcome rather than assuming that successful departmental activity automatically produces a successful business result.

When sales wants a faster promise and operations needs more capacity, who decides what the business should do?

Who sees that the acquisition cost, delivery commitment, refund, and collected money belong to the same transaction?

Who can change the plan when the combined result stops making sense?

You need answers before the next meeting becomes four departments defending four dashboards.

Naming an owner does not erase everyone else's responsibilities. Executives still answer for decisions they retain. Department managers still answer for their work. Employees still need to perform and communicate.

The relationship between those responsibilities needs to be explicit.

One Owner Does Not Mean One Person Approves Everything.

There is an equally stupid way to respond to this problem.

Route every decision through a single executive, then wonder why the business moves slowly.

You have replaced unclear ownership with a bottleneck.

A capable team should be able to act within agreed limits. Specialists should retain the decisions that belong with their expertise. A committee can provide useful judgment without leaving the final decision or follow-through undefined.

The owner of the combined result needs a working decision structure, not a personal invitation to micromanage every employee.

And the business does not need a new title every time two departments disagree.

It needs someone to make the trade-off, record what was decided, and ensure the affected people can act on it.

Collaboration distributes work and knowledge.

It should not dissolve the decision.

Here Is What Would Change My Diagnosis.

Show me that the objective was clear, the authority was appropriate, the necessary information and resources were available, and the handoffs worked.

Show me that risks reached someone who could act, decisions arrived in time, and the team had a reasonable opportunity to deliver.

If all of that existed and the agreed work was still performed badly, individual or team execution may be the problem.

If the work was performed well and the underlying commercial assumptions failed, examine the strategy.

If an external event overturned a reasonable plan, assess how the company prepared for and responded to the event. A bad outcome alone does not establish negligence or incompetence.

For the hypothetical service campaign, I would trace the same jobs from inquiry through promise, delivery, refund, and collection. I would compare the promised scope and timing with what operations knew, what management approved, and the costs that followed.

I would also look at jobs that went well. What was different about the information, decision, or handoff?

If clearer ownership improves the result, track whether the improvement persists and whether it merely shifted costs somewhere else. A better month after a process change is useful evidence, but customer mix, demand, staffing, and other changes can affect the outcome too.

You don't prove the diagnosis by drawing an organization chart.

You test it against the work.

And the review needs to end with a decision: what changes, who can authorize it, who will carry it out, and how the company will know whether it helped.

Otherwise, accountability becomes one more meeting about why the last meeting accomplished nothing.

Follow the Decision. Follow the Work. Follow the Money.

Who chose the objective?

Who approved the assumptions?

Who controlled the resources?

Who made the promise?

Who accepted the next step?

Who could change the plan when it stopped working?

And who was looking at the complete result while everyone else was completing their assigned part?

Those questions do not eliminate uncertainty. They establish where judgment and action belong.

You cannot outsource accountability to the person closest to the failure if the decision that created it was made somewhere else.

If everybody is responsible but nobody can explain who decided what, who had authority, and what success was supposed to look like, the company doesn't have an accountability system.

It has a collection of job descriptions.

Continue Reading on anuhlest

• Who the Fuck Approved This? — why permission to spend is not the same as a justified business decision.
• You Hired Someone. Now What the Fuck? — why responsibility requires a defined assignment, authority, and resources.
• Your Business Isn't Ready for Growth. And That's Your Fucking Problem. — why additional demand cannot repair a broken operating handoff.
• There Are No Five Fucking Ways to Improve Your ROI — why diagnosis must come before prescribing a solution.