Leadership under pressure

The Leadership Gap Nobody Budgets For: Deciding Under Pressure

By Chayill Scott | Dakota Gray, LLC.

The meeting starts with an update.

A deadline is slipping. A client is unhappy. A system is down. A key employee has resigned. Something that seemed manageable yesterday now requires a decision before lunch.

Everyone looks toward the leader.

The information is incomplete. The options are uncomfortable. Waiting has consequences. So does moving too quickly.

This is the moment when leadership stops being a title and becomes a practice.

It is also the moment organizations often discover what they have not prepared for.

They budgeted for the system, the project, the consultant, and the launch.

But did they prepare the people who would have to make difficult decisions when the plan stopped working?

The Gap Between Having Authority and Being Ready

Being authorized to make a decision is not the same as being prepared to make it under pressure.

Consider a manager who handles routine operations well. Priorities are clear. Approvals follow a familiar path. There is time to ask questions and correct mistakes.

Now put that same manager in a situation where two priorities conflict, the facts are changing, and someone wants an immediate answer.

What should they protect first?

What can they decide independently?

When should they escalate?

What happens if their judgment is reasonable, but the outcome is not what anyone wanted?

If those expectations have never been discussed, the manager is not simply making a difficult decision. They are also trying to interpret the organization’s unwritten rules.

That is the leadership gap.

It is the distance between the responsibility people carry and the preparation, clarity, and support they have been given.

Pressure Does Not Resolve Unclear Priorities

Imagine a team facing a service disruption.

One leader says, “Restore operations as quickly as possible.”

Another says, “Do not take unnecessary risks.”

A third says, “Keep the client happy.”

All three instructions sound reasonable. Together, they do not tell the team what to do when speed, risk, and client expectations pull in different directions.

The team needs more than encouragement. It needs a way to make the tradeoff.

Which commitments are nonnegotiable? Which decisions are reversible? Who can authorize an exception? What needs to be communicated before action is taken?

Without those answers, people may delay, seek multiple approvals, or make different decisions from the same set of facts.

The question is not whether they care enough.

The question is whether leadership has created the conditions for sound judgment.

What the Budget Leaves Out

Decision readiness rarely arrives as a neat line item.

It sits inside work that can be postponed because it does not look urgent: clarifying authority, practicing difficult conversations, testing escalation procedures, and working through scenarios before they become real.

There is always another immediate priority.

Until there is an incident.

Then the organization finds time for emergency meetings, executive intervention, repeated explanations, and a review of what went wrong.

The useful budgeting question is not only, “What will leadership development cost?”

It is also, “What are we asking people to handle without preparation?”

A new responsibility should come with more than an announcement. A larger scope should come with more than access to another dashboard.

People need to understand what they are accountable for, what discretion they have, and where support is available.

Decisive Is Not the Same as Fast

Under pressure, speed can look like confidence.

A leader gives an immediate answer. The room moves. Everyone feels relief.

But movement is not the same as progress.

Before committing the organization, a leader needs to separate what is known from what is assumed, identify the consequence of waiting, and understand what cannot easily be undone.

Sometimes the right decision is to act immediately.

Sometimes it is to contain the problem, gather one missing fact, and make the larger decision at a defined time.

Both can be decisive.

What matters is whether the action fits the situation—not whether the leader answers first.

Prepare for the Moment Before It Arrives

Decision readiness does not require predicting every crisis. It requires giving people a reliable foundation for handling uncertainty.

Start with a few practical questions.

What are we trying to protect or accomplish?

What do we know, and what are we assuming?

What can this leader decide without further approval?

What would require escalation?

What is the safest useful action we can take now?

When will we review the decision?

These questions become more useful when teams practice them together.

Take a realistic scenario: a missed delivery, a staffing shortage, an AI-generated recommendation that cannot be verified, or a client request that conflicts with an internal standard.

Ask the people involved to work through the decision.

Notice where they disagree. Notice where authority becomes unclear. Notice which information they need and cannot easily obtain.

Those are not reasons to abandon the exercise.

They are the findings.

Support Cannot Disappear When the Decision Gets Difficult

It is easy to tell leaders, “Use your judgment.”

The harder work is making sure the organization supports them when they do.

If every imperfect outcome produces blame, people have a reason to seek permission for everything. If raising a concern is treated as resistance, concerns may stay unspoken. If escalation leads nowhere, people may stop using it.

Leaders need standards and accountability. They also need a fair distinction between a poor decision process and an unfavorable outcome.

A sound decision can still produce a result no one wanted.

The review should examine what the leader knew, what they considered, whether they stayed within their authority, and what the organization can learn.

Otherwise, the message becomes: “Take ownership—as long as nothing goes wrong.”

That is not much of a leadership invitation.

Build the Capability, Not Just the Expectation

At Dakota Gray, LLC., we help organizations strengthen the leadership infrastructure behind difficult decisions.

That work can include clarifying decision rights, establishing escalation pathways, preparing leaders through realistic scenarios, and creating review practices that turn experience into better execution.

The purpose is practical: leaders who understand their authority, teams that know how to move, and executives who do not have to become the approval point for every difficult situation.

Before the next high-pressure moment, ask:

Where are we relying on individual instinct because we have not built a shared way to decide?

That is a useful place to begin.

Start the Conversation With Dakota Gray, LLC.

If your leaders are carrying more responsibility than their preparation or authority supports, another reminder to “be decisive” will not close the gap.

Contact Dakota Gray, LLC. to discuss a leadership-readiness and decision-making engagement. Together, we can identify where decisions stall, where accountability is unclear, and what your leaders need to act with greater confidence under pressure.

You cannot remove every difficult moment from the work.

You can prepare people to meet it.

Responsible AI

AI Won’t Fail You. Missing Governance Will.

By Melody Fogarty | Dakota Gray, LLC.

Let me ruin the title immediately.

AI will fail you.

Why do I know? Because it is imperfect. So are people. So are organizations. And so, regrettably, are some of the executives announcing that AI will transform their businesses before anyone has established who is allowed to connect it to what.

The title is not a warranty. It is a warning about where leadership responsibility belongs.

An imperfect system will eventually produce an imperfect result. The real question is whether your organization is prepared to catch it—or whether you have created an express lane from an unreliable output to an expensive decision.

One is a technology limitation.

The other is a leadership problem.

At Dakota Gray, LLC., we work on the second.

A Beautiful Answer Can Still Be Wrong

Imagine an AI-generated recommendation that is beautifully formatted, persuasively written, and wrong in the one place that matters.

Who checks it?

Not who could check it. Not who generally keeps an eye on these things.

Who has been assigned to review it, against what standard, before someone acts?

If the answer is a thoughtful silence, you do not need another demonstration.

You need a governance conversation.

We have already spent enough years mistaking confidence for competence in human beings. There is no need to automate the habit.

“We Need AI” Is Not a Strategy

Consider this entirely hypothetical executive exchange.

“We need an AI strategy.”

“Excellent. What are we trying to accomplish?”

“We need to get ahead.”

“Ahead of what?”

“The competition.”

“Doing what?”

“Using AI.”

And there we are: a technology initiative with the strategic depth of a bumper sticker.

A serious AI agenda begins with a business problem, an intended outcome, and a clear basis for investment. It identifies who owns the work, what success looks like, and what would justify stopping.

It also leaves room for an unfashionable question:

Could we solve this more effectively without AI?

If that question irritates the sponsor, we may have discovered that the initiative is less about organizational performance and more about appearing innovative.

An expensive distinction. Often billed monthly.

Dakota Gray helps executive teams turn enthusiasm into a disciplined agenda: where AI belongs, what deserves investment, and what the organization must have in place before proceeding.

Governance Is Where the Glamour Gets Its Instructions

Governance is not a committee assembled to admire a policy.

It is the practical structure that answers the questions leadership cannot afford to leave unresolved.

Who can authorize an AI use case? What information may it access? Which outputs require human review? Who can pause deployment? How are problems escalated and resolved? Who decides whether an initiative should scale—or stop?

These questions belong before implementation, not in the emergency meeting afterward.

Curiosity is welcome. Unsupervised organizational exposure is not a professional-development plan.

Good governance makes responsible action possible without depending on everyone’s personal interpretation of “probably fine.”

That is why our work connects strategy to decision rights, operating procedures, and executive oversight. A policy is useful. A policy someone knows how to execute is considerably more useful.

Empathy, Not Sympathy

There is a human side to this work. I mean empathy—not sympathy.

Sympathy can remain comfortably outside someone else’s experience.

“Oh, this transition must be difficult for you.”

Empathy asks leadership to understand what the transition actually requires of people.

What changes in their work? What new judgment will they need? Can they challenge an AI-generated recommendation? Will they be supported for raising a concern—or treated as an obstacle to progress?

Suppose an employee is expected to review AI output without adequate time, training, or authority to reject it.

Calling that arrangement “human oversight” does not make it meaningful.

It makes the employee your designated recipient of consequences.

Elegant terminology. Unattractive management.

People do not need a sympathetic speech followed by an impossible assignment. They need a workable operating model.

At Dakota Gray, we help leaders design human-and-AI work around clear roles, realistic expectations, review requirements, and escalation authority. Adoption should not depend on employees quietly absorbing the gaps in the implementation plan.

Human in the Loop. Authority in the Room.

“Human in the loop” sounds reassuring.

Which human?

With what expertise?

At what point?

And can that person actually stop the process?

A reviewer who can identify a problem but cannot intervene is not an effective control. They are a witness.

Executive oversight deserves the same scrutiny. Someone must maintain a clear view of the AI portfolio: what is being funded, what is being deployed, what is producing value, and what is becoming an obligation nobody intended to own.

A successful demonstration is not permission for unlimited expansion.

Someone still needs to assess the costs, dependencies, review burdens, and ongoing responsibilities.

Innovation does not become mature because the presentation received applause.

The Technology Does Not Own the Decision

When an AI initiative goes wrong, “the system recommended it” explains the sequence. It does not transfer responsibility.

Someone authorized the use case.

Someone approved the access.

Someone accepted the review process.

Someone decided the organization was ready.

That is where accountability belongs.

The executive task is not to promise that AI will never make a mistake. It is to establish how mistakes will be detected, contained, corrected, and used to inform the next decision.

You do not need perfect technology to exercise sound leadership.

You do need leadership willing to ask inconvenient questions before the answers become expensive.

Is Your Organization Ready—or Merely Equipped?

Before approving the next AI initiative, ask your executive team three questions.

Can we identify every active AI initiative and its accountable owner?

Can we explain the approval, review, and escalation requirements for each?

Can we show whether the investment is delivering value after accounting for implementation, oversight, and ongoing support?

If those answers are incomplete, that is the place to begin.

Not with another tool.

With the leadership infrastructure around it.

Start the Conversation With Dakota Gray, LLC.

Dakota Gray helps executive teams establish AI strategy, governance, investment priorities, and human-and-AI operating models—from readiness assessment through implementation oversight.

We help create the structure to lead an AI portfolio, oversee technical deployment alongside qualified specialists, prepare people for changed work, and evaluate organizational results.

Whether you are assessing your first use case or trying to bring order to initiatives already underway, the starting point is a focused executive conversation about your priorities, readiness, and oversight gaps.

Contact Dakota Gray, LLC. to discuss an AI readiness and governance engagement—and establish what needs to be in place before your next investment, pilot, or expansion.

AI will fail you occasionally.

Missing governance is how you give that failure permission to become everyone else’s problem.