Data Quality, Trust, and AI Readiness—Lessons from Tom Redman

In this special edition of The Executive Outlook , we had the privilege of speaking with Tom Redman , a man better known across the world as The Data Doc. For decades, Tom has been a trusted voice in the data community, known not only for his practical wisdom but also for his ability to explain complex issues in ways that make sense to everyone, from analysts on the ground to executives in the boardroom.

His articles in Harvard Business Review and his book People and Data: Uniting to Transform Your Organization have reached thousands of practitioners, giving them both inspiration and tools to take charge of their data challenges. 

But the story of how Tom came to be The Data Doc isn’t about titles or self-promotion. It started with his work excellence. 

Tom chuckled as he recalled it. “We were deep into a project, working through tough questions about how the organization was handling its data. At one point, the client looked at me and said, you’re like a data doctor. I didn’t think much of it, but I decided to try it out on a business card. People instantly resonated with it. It carried a sense of trust. If you’re a doctor, people believe you’ll help them heal their problems. That simple nickname became a cornerstone of how people saw me.” 

From that moment, Tom leaned into the idea. And for years since, the Data Doc has been helping organizations treat the real root causes of their data illnesses. 

Tom has worked with organizations of all shapes and sizes, fast-moving startups, century-old giants, oil companies, banks, tech innovators, and consumer brands.

Prefer to listen on the go? Tune in to the full podcast episode on Spotify below:

Yet when asked about the most common challenge he sees, his answer came without hesitation: companies are far too tolerant of bad data. 

“Every company has data quality problems,” he explained. “But what worries me is how much tolerance there is for them. People accept errors as if they’re normal. One department makes a mistake, another department spends hours fixing it, and everyone carries on as if that’s just the way things work.” 

He pointed out how damaging this mindset can be. “In many organizations, employees spend half their day, sometimes more, fixing data issues before they can even do their real work. Think about that. Highly skilled professionals, people trained in finance, marketing, or engineering, are spending a huge chunk of their energy cleaning up other people’s messes. Who really wants a career like that?”

Tom’s concern isn’t just about wasted time. He sees a direct connection between this tolerance for poor data and the growing pressures around artificial intelligence. “AI depends on solid data foundations. But if your data is weak, you can’t trust the answers. And ironically, the more important the decision, the less you can rely on the results if the data quality is bad. That’s dangerous.” 

Despite this warning, Tom isn’t pessimistic. He believes in AI’s potential. “I want AI to succeed,” he said with conviction. “It has the ability to solve massive problems in business and in life. But we’ve been through AI winters before—times when excitement turned to disillusionment. And bad data could easily trigger another one. People have been patient with mediocre AI so far, but trust is fragile. If organizations continue to feed poor data into their systems, AI will not live up to the hype.”

His prescription is clear: fix the foundation or risk the collapse of trust.

Watch the full conversation on YouTube by clicking the link below:

Throughout his career, Tom has worked tirelessly to show that data quality is not an abstract issue. He has documented real-world success stories from companies like AT&T, Chevron, Shell, Gulf Bank, HelloFresh, and Era Energy. Many of these case studies have been published in respected journals, giving the entire industry examples of what’s possible. 

“What I want is for data professionals to put their success stories into the public domain,” Tom explained. “When people see what’s possible, when they see companies saving millions, reducing risk, or building trust, they realize data quality isn’t just about fixing numbers in a spreadsheet. It’s about enabling the business to function with confidence.”

One of his favorite statistics comes from McKinsey: employees spend about 30% of their time dealing with data issues. “That’s not my number, though it sounds like one I’d come up with,” Tom joked. “But think about it. It means almost every job has two parts: doing the work and fixing the data so you can do the work. And most people never signed up for that. They studied finance, marketing, or computer science, not cleaning up crap every day. It drags down morale, it kills productivity, and it stops people from doing what they’re truly capable of.” 

What excites Tom most is moving organizations from this frustrating cycle of error-fixing into a place of empowerment. “People love solving root causes. They love working together, eliminating problems at the source, and building better systems. It’s not just about saving money; it’s about giving employees the satisfaction of making a lasting difference.” 

That theme of empowerment is at the heart of Tom’s message. He believes that improving data quality isn’t just a technical issue—it’s a cultural one. It’s about unleashing the energy of people so they can focus on creating value instead of patching mistakes.

When asked what advice he would give to organizations just starting out, Tom Redman gave two pieces of guidance. 

The first is to start with a big, visible business problem that’s rooted in bad data. “At AT&T, we focused on invoicing errors. People didn’t trust the invoices, and it created massive inefficiencies. Once we fixed it, hundreds of people’s workloads improved immediately. That built momentum and showed leadership that solving data problems makes a difference.” 

The second is to gather facts. “Everyone has opinions about whether their data is good or bad. Without evidence, it’s easy to shrug and say we’re no worse than anyone else. I created something called the Friday Afternoon Measurement Method. The idea is simple: pick a process, look at the last hundred times it was done, and count the errors. It only takes a couple of hours, but it gives you hard data. If you do this across 50 key processes, you suddenly have a map of your organization’s data quality. That kind of evidence changes the conversation.” 

Tom also reflected on common mistakes he sees in analytics projects. “The biggest one is skipping the step of defining the problem. Analysts often want to dive straight into the data, but without agreement on what you’re solving, you risk chasing the wrong thing.”

He leaned on an Einstein quote to make his point: “If I had an hour to save the world, I’d spend 59 minutes understanding the problem and one minute solving it.” For Tom, that wisdom applies perfectly to data work. “Clarity upfront saves you from wasted effort and bad conclusions. It’s the most important step, yet it’s the one people overlook the most.” 

Looking back over his decades of work, Tom reflected on how the field has evolved. “The overall quality of data isn’t much better or worse than it used to be. But what has changed is its importance. With AI, unstructured data, and global complexity, the stakes are higher. Data quality is no longer a side issue—it’s central to whether organizations succeed or fail.” 

He credited pioneers like Danette McGilvray, Larry English, Richard Wang, and others for laying the foundations of modern data quality practices. Together, they have shaped approaches that move organizations from constantly fixing errors to preventing them at the source. Still, Tom admitted, one challenge remains unsolved: “We haven’t yet made data quality mainstream. AI is cool. Data quality should be cool too. But we’re not there yet.”

On the topic of tools and governance, Tom was candid. “Most data quality tools promise too much. They help you find errors faster, but they don’t eliminate the causes. Unless you attack the root, you’re just applying Band-Aids. Tools have their place, but they don’t solve the problem.” 

What works better, he argued, is leadership. “Executives want to help, but they often don’t know how. Don’t try to explain every technical detail to them. Instead, show them exactly where they can contribute—like building a network of embedded data managers. Make it simple for them to take action. That’s when you get real progress.” 

As our conversation drew to a close, Tom shared a message that was both urgent and hopeful. “Improving data quality is about more than saving money or complying with regulations. It’s about trust. It’s about empowering people. It’s about creating workplaces where employees stop cleaning up crap and start creating real value.” 

For Tom Redman, the Data Doc, the prescription is clear: become intolerant of bad data, fix the root causes, and put people at the center of every solution.

His closing words carried the weight of a lifetime of experience: “Don’t wait for perfection. Start with one problem. Solve it. Build momentum. Data quality isn’t just a technical issue—it’s the foundation for better business, better AI, and better work lives.”

Click here to discover more life stories and insights from leaders shaping the future of data and tech.