REPLACE Big Data Show logo
April 30 - May 1, 2025
North Javits Center | New York City

Talking Data with Eric Kavanagh

Host of DM Radio and Program Advisor to Data Universe

In less than six months, Data Universe will be holding its inaugural event in New York City, April 10-11, 2024.  

When we decided to build this event for the US, which will take place at the North Javits Center—a sparkling technologically advanced venue on Manhattan’s west side—one of our core objectives was to find and deliver the most relevant, up-to-date, and compelling intelligence available pertaining to the data ecosystem, in a vendor-neutral format. To ensure we are aligned with market needs on this mission, Data Universe has assembled an exceptional network of global experts from different areas of the data space to provide advice and guidance on our conference topics and event experience. Leading up to Data Universe in 2024, we will be highlighting each of our advisors and their careers, industry insights, and thoughts about the unique potential for this event, as well as their role in it.

Since 2008, Eric Kavanagh has been the host of DM Radio, streaming online and broadcasting over the air in more than a dozen national markets each week. It is one of the longest running and best-known shows devoted exclusively to data as a business resource. His mission: have ongoing conversations with the best minds in industry and academia about data and all the ways businesses can leverage it.

Data Universe: Data is a broad topic and getting broader. Are there fundamental themes you want to make sure you hit each time you broadcast?

Eric Kavanagh: It changes weekly based on the topic. What we want to make sure of most is that we have an unscripted, but guided conversation. When you get several experts in the same domain on the same show they get excited and our listeners benefit from that. In the early 2000s I understood the potential for data, but at the time I didn’t see where the nexus would occur between data and AI. And now that has become the bottom line. Whatever you want to do with AI, to train it, to use it, to execute it, you need data as the foundation. So, our mission is to get people to understand it.

DU: You mentioned AI as the topic that’s binding everything together right now. But is AI sucking all the oxygen out of the room when it comes to discussions around data. ? Should we be talking more about other areas or is AI just that important?

EK: Both. It is absolutely sucking the oxygen out of the room. LLMs, AI and the foundational models are absolutely dominating the conversation. There’s been $30 billion in investment in AI related to LLMs in less than a year. But it is that important. think there isn’t any use case that’s not in the crosshairs right now for AI. Marketing and content creation are dominating now, but it’s just the beginning. Interactive AI will take that to a new level. The complexity of these models learning to reflect back fully syntactically correct prose is pretty amazing. But if you think about machine-to-machine conversations, they’re a lot simpler. In my opinion, that means a foundational model should be very powerful in terms of optimizing your current information and systems architecture. I think we’re going to see crazy efficiency gains using these foundational models and AI to get what the old data architecture does way less efficiently.

DU: What is the state of data literacy for the average business user? Are they ready to get the most out of the advancements you see coming?

EK: I think the Pareto Principle—the 80/20 rule—applies here. But I think you have to break it down further than that. Twenty percent of the 20 percent—about four percent—are all in. They totally get data, they’re working with it, they’re focused on it and they’re excelling. The rest of the 20 percent are open to it, they’ve tested it out a little but their knowledge is surface. And 80 percent are just too busy doing other stuff. To quote a friend of mine, the “tyranny of urgency” is driving the reluctance or the inertia involved in leveraging data technologies. But as more AI technologies are embedded in Microsoft and Google products more people will have exposure to data and AI technologies. Then it becomes a question of how useful are they, have they sorted out the guardrails, etc.

DU: You’re a broadcaster interested in taking your message to as many people as possible. But how important are face-to-face industry events for advancing the general expertise of data and analytics?

EK: They’re hugely important and they have come roaring back since Covid. I’m a virtual guy, but virtual will only get you so far. But these focused events are crucial. The broader, consumer-based events don’t allow you to have focused conversations in your area. I’m a data guy, so these events are the best places to have these conversations and where you can learn the most.

DU: What do you think will be the most important takeaway for people who attend Data Universe?

EK: The granular knowledge you can pick up at a well-designed event like this is so valuable. I love round tables and I love to moderate round tables because at their best there is a group chemistry and a momentum that enables people to learn more from each other than they might in a session with a different format. And the free food and cocktails!

DU: As someone advising on the program what do you hope this event will deliver to the data community?

EK: Lots of hard facts. When we’re developing the program, we want to make sure we are delivering the straight and honest truth about the technologies and what they’re capable of. When speakers and presenters know there’s someone in the room who knows all their weaknesses, boy do they behave. From my experience on my show, I know if I have an analyst and two vendors the conversation becomes very constructive. Executives from large software companies are like politicians. They have their messaging down and you have to find some way to get to the gritty details. I want to be friendly, but I want to get to the meat. I want to help design a conference that provides the meat.