I’ve Seen Every Enterprise Technology Revolution Since the 1990s. The Biggest Mistake Never Changes.

I have spent nearly thirty years recruiting into enterprise technology.

ERP was going to change everything. Then CRM. Then cloud. Then digital transformation. Then automation. Now AI.

Every wave arrived with a different acronym and the same promise. And every wave produced the same failures, which were almost never technology failures.

Businesses rarely fall behind because they picked the wrong platform. They fall behind because nobody decided who was responsible for making it work.

That has been true for thirty years. I do not think AI has changed it. I think AI has raised the price of getting it wrong.

The same mistake keeps coming back

A business spends eleven months choosing a platform. Countless meetings. Endless demos. A procurement process that outlives two of the people who started it.

Then it goes live, and everyone waits for the organisation to transform.

It has never worked like that.

I have watched brilliant platforms underdeliver because nobody owned adoption, nobody led it visibly, and nobody changed how the business actually worked around it.

I have also watched businesses get extraordinary results from unremarkable technology, because the right people were behind it and they started early.

The platform was never the variable. It just happened to be the thing on the invoice.

Recruitment gives you a different view

You are never watching one transformation. You are watching dozens at once, at dozens of different stages, across businesses running the same programme. It gives you something no single company can buy, which is a longitudinal view.

It also gives you an early one, because hiring is the leading indicator. A company announces a strategy in January and it means nothing. A company opens a requisition in January and it means everything. The requisition has cleared finance.

A job specification is the most honest document a company produces. Not the strategy deck. Not the town hall. The spec, because that is where an organisation writes down what it is willing to pay for.

And what I see, over and over, is the same shape. The businesses that succeed hire the leadership before the programme becomes business critical. They do not wait for it to struggle and then look for someone to rescue it.

They think ahead.

Decades of research says the same thing. Prosci has consistently found that active and visible executive sponsorship is the single biggest contributor to successful organisational change. Not the technology. The people who bring the organisation with them.

AI has not rewritten the rules. It has raised the stakes.

AI does feel different. The pace is extraordinary, capabilities move weekly, and the pressure to act is real. But the difference is not that the technology is cleverer. It is that the technology now acts.

An ERP system sat there and waited for a human to use it properly. An AI agent reads the ticket, calls the API, and closes the record at 3am on a Sunday while nobody is watching.

The old question was: will people adopt this?

The new question is: when it acts, who answers for it?

That is not a technology question. It is the same people question it has always been, arriving with a deadline attached.

Which is why the conversation has moved from automation to orchestration. The value is not in having intelligent systems. It is in knowing which system is permitted to act, under whose authority, with what audit trail, and what happens when it gets something expensively wrong.

That is close to what McKinsey found in its latest State of AI research. The businesses seeing the greatest value are not simply adopting the technology faster than everyone else. They are redesigning workflows, strengthening governance, and putting leadership structures in place to hold the change.

Almost every organisation I speak to tells me governance is a priority. Very few have put a name and a salary band against it. The gap between those two facts is the whole story.

Technology eventually becomes available to everyone

I have seen enough cycles to know that today’s advantage rarely stays exclusive. The platforms converge. They always do, and with AI they are converging faster than anything I have watched before. So what separates companies is not the technology. It is what they built around it.

The businesses that get the most from AI will not be the ones spending the most. They will be the ones who can put a person in front of a board, or a regulator, and have them explain a decision their software made without asking anyone.

You do not hire that person the week you need them. You hire them now, or you inherit their absence.

Some lessons do not date

Thirty years on, the technology has changed beyond recognition. I have watched platforms come and go, acronyms take over the headlines, and every generation arrive convinced it is living through the biggest shift yet. In fairness, most of them were right.

But what makes a transformation stick has never moved. It is not the software. It is not the implementation. It is not being first.

It is having leaders who give people confidence, explain the bigger picture, and help the business adapt as it goes.

Technology creates the opportunity. People turn it into lasting change.

I would add one thing to that. “It is about the people, not the technology” was always meant as a warning. Lately I hear it used as reassurance, and it has started to sound like something executives say instead of assigning ownership. It sounds wise. It costs nothing. It commits no one.

So let me put the question to the room, because I would rather be corrected than agreed with.

In your organisation, right now, is there a single named human who is accountable when an AI system takes an action that turns out to be wrong?

Not a committee. Not a policy. A name.

If there is, tell me what the role is called. The market has not settled on a title yet, and I am tracking it. If there is not, you are in the overwhelming majority, and that is the more interesting answer.

This is the question I keep coming back to in Now Intelligence, my newsletter for the ServiceNow community. Whether I am writing about AI governance, enterprise architecture or the future of orchestration, it is always the same one: how do organisations turn technological progress into meaningful business change?

The technology evolves.

The leadership challenge does not.