If AI Can Do This Now, Why Hire an Operations Consultant?

“Hey ChatGPT, analyze my company.”

That question is getting more reasonable every month.

AI can:

• help write SOPs;

• recommend workflows;

• summarize and analyze meetings;

• identify automation opportunities;

• help design and configure systems.

And depending on the tool, it can increasingly help build parts of the operating environment too.

So if you're a CEO, the obvious question is:

Why would I still pay someone to help with operations?

I think that is a fair question.

And I also think parts of operations consulting are absolutely becoming commoditized.

That’s probably a good thing. A business shouldn’t pay for hours of manual work when AI can produce a strong first draft in minutes.

The value is shifting.

Good operational work is partly about finding the missing information

This is where human judgment becomes more valuable in an AI-enabled company.

The work is closer to:

• What do we think is happening?

• What is actually happening?

• Where are those versions different?

• What information is missing?

• What is the symptom versus the cause?

• What should be fixed first?

Companies rarely need everything fixed at once.

They need the right next operating improvement.

AI can generate twenty excellent recommendations. Judgment determines which two are worth implementing now.

Producing the answer is becoming the easy part

Imagine you tell AI:

“Our company is too dependent on me. We need better systems.”

You may get a very good response.

• Clarify ownership.

• Document critical processes.

• Establish decision rights.

• Improve visibility.

• Automate repetitive work.

The problem is that they depend entirely on the accuracy of the problem you've described.

• What if the manager already has authority—but doesn’t believe leadership will support the decision?

• What if the process is documented but nobody uses it?

• What if the software already has everything the company needs, but the workflow was never designed across departments?

• What if everyone agrees on what is supposed to happen—but the actual process contains undocumented workarounds?

AI can reason brilliantly from the information it has.

But someone still has to determine whether that information represents reality.

The harder question is: Are we solving the right problem?

A CEO may say:

“We need better documentation.”
Maybe. Or perhaps the process is documented and nobody uses it.

“We need new software.”
Maybe. Or perhaps no one has defined what the software needs to support.

“I need to delegate more.”
Maybe. Or perhaps responsibility moved without enough authority or context.

AI can help solve any of those problems.

But a sophisticated solution to the wrong problem is still the wrong solution.

AI can help you move faster.

Which means it can also help you move faster in the wrong direction.

AI needs context—and businesses are full of missing context

Strong operational recommendations require accurate context:

• how work actually moves;

• who owns and decides what;

• which systems people actually use;

• where work stalls or workarounds appear;

• what happens when the normal process breaks;

• where information gets lost between functions;

• and what leadership genuinely needs visibility into.

That information rarely exists in one clean document waiting to be uploaded.

It lives across software, meetings, spreadsheets, email, SOPs, individual memory, and the people doing the work.

And sometimes those sources contradict one another.

It may be the most useful information you find.

“We already documented that.”

Here's a simple example.

A CEO says:

“We already documented that process.”

Okay.

Where does the team go when they need it?

“It's in SharePoint.”

Do they actually use it?

“Probably not.”

Now we know the issue may not be documentation.

• It may be adoption.

• Accessibility.

• Or a gap between the documented process and how the team actually works.

AI could absolutely help once it knows that.

But someone had to ask the question first.

The sequence matters

Suppose leadership wants automation.

Great.

But if ownership isn’t clear, automation may simply automate ambiguity.

Suppose the company wants an AI knowledge system.

Great.

But if the source material is inconsistent, outdated, or contradictory, you’ve made bad information easier to retrieve.

Human judgment doesn't mean avoiding AI

I use AI because it can make operational work dramatically faster.

It can synthesize information, draft documentation, compare processes, surface inconsistencies, generate workflow options, and accelerate repetitive implementation work.

I don't think a client should pay me to manually perform work an AI-assisted process can complete faster.

But the question I keep coming back to is:

What should still require judgment?

• What is actually worth fixing?

• Which process should be redesigned rather than documented?

• Which decisions genuinely require executive judgment?

• Which workaround is waste—and which one is revealing something important?

• Which recommendation looks efficient but won’t survive contact with the team?

Those are operating decisions.

And those decisions become more—not less—important as implementation gets faster.

If your company is trying to use AI, software, automation, or better systems to improve execution—but you're not yet sure what should be fixed first—the Digital Operations Assessment is designed to identify the underlying operating priorities before more technology gets layered on top.

The solution still has to survive contact with the organization

Even the right solution can fail if it doesn’t fit how people actually work.

Someone still has to test it in reality, observe where friction appears, adjust the design, and reinforce how the system is used.

AI can support that work too.

A quick diagnostic for leaders using AI in operations

If you're wondering whether AI can solve the problem on its own, start here:

1. Can you clearly explain the operational problem without describing only its symptoms?

“We're always behind” is a symptom.

So is “everything comes back to me.”

If you don't yet know what is causing those outcomes, diagnosis needs to happen before solution design.

2. Does the information you're giving AI describe how work actually happens—or how leadership believes it happens?

Those can be very different.

3. Do you know what information is missing?

AI can identify gaps in the information it sees.

It cannot always know which important context was never provided.

4. Are different people describing the same process differently?

That's not noise.

That contradiction may be the discovery.

5. If AI gave you ten excellent recommendations today, would you know which one the organization should implement first?

Prioritization is part of the operating problem.

6. Who will make sure the solution works once it meets real people?

Someone still has to test it, observe what happens, gather feedback, adjust the design, clarify expectations, and reinforce adoption.

So why hire an operations consultant in an AI-enabled world?

Not because AI can't write the SOP. It can.

Not because AI can't recommend a workflow. It can.

Not because AI can't help build the system. Increasingly, it can.

The value is determining:

• what should actually be solved;

• what context matters;

• what information is missing;

• what should happen first;

• and whether the solution works once people start using it.

If you have the time and operational perspective to investigate the business, find the contradictions, talk to the right people, diagnose the underlying problem, prioritize the solution, architect the environment, and lead adoption—

you may not need an operations consultant.

You've basically become one.

When producing the answer gets cheaper and faster, the advantage shifts to knowing which problem deserves the answer.