Stephen Gatchell explains why AI strategy and consulting must align people, data, governance and business outcomes before technology can scale.
During a recent conversation with The Executive Outlook, Stephen Gatchell shared a pattern he has seen more than once. A company asks for help with its AI strategy. The conversation starts with technology, but before long, it becomes clear that technology is not really a problem.
Maybe the roles are unclear. Maybe leaders are not aligned on what they want AI to achieve. Maybe employees do not know how their work is going to change. The AI conversation is often just where the deeper problem first becomes visible.
Stephen is Partner and Head of AI Strategy at Ortecha. He has spent more than thirty years working across technology and data, seeing the industry from several sides. He has worked inside large organizations, advised hundreds of companies, sold technology as a vendor and now helps businesses turn data and AI into something useful.
He also holds a patent related to managing company data.
What stands out in his perspective is not another prediction about where AI is going. It is the way he keeps bringing the conversation back to a simple question. What is the business actually trying to accomplish?
That question sits at the heart of his approach to AI strategy and consulting.
One of his first jobs was banking customer service.
At the time, data was nowhere near as central to business conversations as it is today. AI was not part of the picture either.
But he noticed something early.
Good customer service depends on the information.
Who is the customer? What do they need? What problems are they dealing with? What are you already doing for them?
When you know those things, the conversation changes. Customers feel understood. Trust becomes easier to build.
Later, He moved into technology, where he worked on development and management reporting. That gave him a different view of the same idea.
People were making important decisions using the data his team provided.
That meant the data had to be accurate, valid and relevant.
The roles have changed over the years. Customer service. Sales. Application engineering. Technology.
The common thread did not.
It was always data.
Stephen has worked inside large enterprises, smaller organizations, technology vendors and consulting firms.
Each experience changed the way he looks at AI transformation.
A company deciding how to use AI has to understand more than the technology itself.
It needs to know whether a vendor can become a real partner rather than simply sell another platform.
It needs advisors who understand the business problem behind the technology request.
And it needs a clear view of its own organization, including reporting structures, responsibilities, operating models and the impact that change will have on employees.
That is where AI strategy and consulting become broader than choosing tools.
Technology decisions live inside organizations. They affect people, processes, budgets, customers and leadership priorities.
A strategy that ignores those connections may look impressive on paper and still struggle in practice.
Stephen has seen companies bring Ortecha in for what they describe as an AI strategy problem.
Then the conversation begins.
Soon, the focus moves away from AI itself.
The real questions become much more human.
Who owns what? Are responsibilities clear? How will someone's day to day work change? What new skills will people need? Does the organization have the structure to support what it is trying to build?
This is why he comes back repeatedly to the idea of being human first.
It may sound simple, but the reasoning is practical.
It is easy for leadership teams to become absorbed in models, platforms, agents, and automation. Employees experience the change differently.
They may be wondering whether their job will still exist. They may be unsure whether they have the right skills. They may not understand what AI means for the work they do every day.
Those concerns are not separate from AI adoption.
They are part of it.
Putting people first does not mean making technology less important. It means making sure the organization is ready to use that technology well.
When he talks about leadership in data and AI, he describes a simple triangle.
Corporate strategy sits at the top.
Technology sits on one side.
People sit on the other side.
The job of the leader is to connect all three.
That sounds straightforward, but it requires something many organizations still struggle with. Leaders have to understand technology well enough to know what is possible, understand the business well enough to know what matters and communicate clearly enough to connect the two.
He sees one particular problem repeatedly.
Companies are investing in AI and managing data, but those efforts are not always clearly connected to corporate strategy.
When that connection is missing, sponsorship can disappear.
Executives need to understand how an AI initiative supports the goals they are already responsible for delivering.
That is why alignment must come before execution.
Before choosing another platform or launching another AI program, leadership should be able to answer a basic question.
How does this help us achieve what the business has already decided on matters?
Technology comes after that.
Governance does not usually create excitement in a leadership meeting.
Stephen understands that.
Rather than treating governance as another compliance exercise, he frames it around what it enables.
If a company has strong data foundations, follows established frameworks such as NIST or DCAM, and invests properly in how data and AI are managed, governance can make adoption easier.
Teams know what they can do.
Responsibilities become clearer.
Risk is easier to understand.
Decisions move with more confidence.
That is a very different conversation from telling the organization it needs more rules.
Governance works best when people see it as an infrastructure for progress, not simply another layer of control.
Another idea he challenges is the belief that every piece of data needs to be perfect before an organization can do anything meaningful with AI.
His view is more practical.
Data should be fit for purpose.
Start with the use case.
Understand the decision the data needs to be supported.
Then decide what level of quality is actually required.
Not every data set carries the same risk. Not every business decision requires the same standard.
That does not mean data quality matters less.
It means quality has to be understood in context.
AI can sometimes work around weaknesses in data, but only when the organization understands the use case and knows what level of quality is acceptable for that situation.
For leaders, this can change the way data investments are prioritized.
Instead of trying to make everything perfect, teams can focus effort where poor quality would genuinely affect an important decision or outcome.
Stephen has also seen another major shift during his career.
Data is no longer mainly rows and columns.
Documents, PDFs, conversations, external information, connected devices, and other forms of unstructured data are becoming increasingly important.
As more sources come together, context becomes just as important as the data itself.
A system does not only need information.
It needs to understand what that information means and how different pieces relate to one another.
Stephen sees an interesting difference between organizations at different stages of AI adoption.
Some companies are already using unstructured data without thinking much about it.
They use tools such as Microsoft Copilot, and those tools work across documents and other forms of enterprise information in the background.
More mature organizations are going further.
They are building agents and exploring increasingly autonomous processes. At that point, unstructured data can no longer be something that technology simply happens to access.
It must become something the organization deliberately manages.
That creates a useful question for CIOs and data leaders.
Is your unstructured data simply being used, or is it actually being managed?
Those are two very different levels of maturity.
He also believes the amount of stored data will continue to increase significantly.
The impact is already moving beyond the software. The growth of AI is influencing data center expansion, infrastructure requirements, and energy demand.
AI may feel digital, but the infrastructure supporting it is very physical.
Some of the companies he works with are not simply using data internally.
They sell data products.
That introduces a different set of questions.
How can AI help create new products?
Can it improve processes that are still manual?
Can quality problems be identified before customers are the ones reporting them?
The answers extend far beyond engineering.
The organization must think about responsibilities, risk, compliance, audits, customer delivery, maintenance, testing and future product development.
The goal is not simply to build a good data product.
It is to understand how that product creates revenue and how trust in it is maintained.
AI then introduces another commercial question.
Can the company build more products or improve existing ones faster because AI is now part of the process?
At that point, AI is not just a technology discussion.
It becomes part of the business model.
When he is asked about the work he is most proud of, he makes an important correction.
For him, it is not about personal achievement.
It is about the team.
He points to his time running engineering labs at EMC.
When he took over, customer meetings were often dominated by complaints.
Expectations were being missed and the people working in the labs were struggling with the environment.
Over time, that changed.
One part of the solution was people.
The roles became clearer. Employees received training. The team worked on communication, customer interaction and change management.
These were not highly technical improvements, but they mattered.
The second part of the solution came from data.
The team used information from its ticketing system to identify when resource problems were likely to occur.
Instead of discovering the issue after a deadline was missed, they could see it coming and manage expectations earlier.
The lesson is simple.
The transformation worked because they did not treat it as only a people's problem or only a data problem.
They addressed both together.
The culture changed alongside the infrastructure.
Stephen's view of responsible AI is grounded in everyday business reality.
It means using AI efficiently.
It means understanding the cost of using it.
It means thinking about energy and infrastructure.
And it means working within internal policies and regulatory expectations.
His definition makes responsibility part of normal operational decision-making rather than a separate conversation that happens after the technology has already been deployed.
There is another decision he believes leaders need to make.
What should AI be allowed to do on its own?
What should remain semi-autonomous?
Where should people stay directly involved?
There is no single answer for every company.
The decision depends on risk, context and what happens if something goes wrong.
But leaders do need to make those boundaries clear.
And then they need to explain them to the business, technology teams and employees.
Near the end of the conversation, he brings the discussion back to something very practical.
Start with what you are trying to do.
Why did you buy technology?
Why are you changing the process?
What outcome are you expecting?
Those questions cut through much of the noise surrounding AI.
He also warns organizations not to try to change everything at once.
Choose a specific use case.
Deliver an outcome.
Learn from it.
Then move to the next one.
Declaring that a company with tens of thousands of employees will suddenly become AI first does not create value by itself.
Leadership still has to define what that actually means.
The same is true of complexity.
AI can be complicated, but not every useful AI project needs to be.
Many companies are still at the beginning of their journey. Their most valuable first-use case may be surprisingly simple.
There is no benefit in making it bigger simply because technology can do more.
He also places a great deal of importance on AI and data literacy.
If a company says it wants to become AI first, the people inside the organization need a shared understanding of what that means.
Otherwise, everyone is moving toward a different version of the same goal.
The pace of technology creates another problem.
Companies can easily fall into a cycle of constantly changing platforms, processes, and roles.
Stephen's warning is simple.
People get tired of constant change.
Organizations need enough stability to actually execute.
Choose the process.
Choose technology.
Deliver something measurable.
Then build on what works.
There will be times when technology needs to change. Business strategy changes. Requirements change. Better capabilities emerge.
But changing direction every time a new platform becomes popular can create more disruption than progress.
The objective is not to own the newest technology.
It is to produce the outcome of business needs.
After more than three decades of data and technology, Stephen's message is surprisingly simple.
Too many professionals focus on technology and not enough on people and outcomes.
That matters because AI is no longer limited to technology companies or a small group of early adopters.
Retailers, healthcare organizations, nonprofits, financial institutions, oil and gas companies and businesses across other industries are all trying to understand how their data and AI capabilities can create an advantage.
Technology has changed dramatically.
The leadership question has not.
What are you trying to accomplish?
Are your people ready for the change?
Does technology support strategy?
That is where AI strategy and consulting have real value.
Not in making AI sound more sophisticated, but in connecting technology to people, business decisions, operating models and measurable outcomes.
Get those connections right and technology becomes an enabler.
Miss them and even the most advanced platform will struggle to create meaningful value.
Are you a CIO, CDO, data leader, AI executive, or technology founder with experience that can help other leaders make better decisions?
Share your perspective, leadership lessons and industry insights with an executive audience.
Get featured in The Executive Outlook and join the conversations shaping the future of data, AI and digital transformation.
Listen in for business strategies and interviews with business experts to help you build a business and a life that you love.
About Us
