Tamr’s 5 Key Takeaways from the Gartner Data & Analytics Summit

After attending the 2025 Gartner Data & Analytics Summit in Orlando earlier this month, it’s hard not to feel inspired. Billed as “the premier location for data, analytics, and AI leaders to uncover the latest in data management, data trends, governance, and data architecture to deliver value for the future,” this year’s summit did not disappoint. With dozens of sessions across five tracks, there was certainly something for everyone, regardless of where you are in your data and analytics career. 

As we looked across the sessions, however, five key themes caught our attention. And from what we heard, they are among the many topics keeping data leaders up at night:

• AI – and AI agents in particular – received a lot of airtime
• Technology investments need a direct tie to business value
• “Boil the ocean” is not a viable strategy (as if it ever was!)
• AI literacy is mandatory, not nice-to-have
• Trustworthy data is critical, but difficult to obtain

AI – and AI Agents in Particular – Received a lot of Airtime

Artificial intelligence (AI) has been the hot topic for many years now. But this year’s conversation went in a different direction: agentic AI. According to Forbes, agentic AI is “artificial intelligence systems that possess a degree of autonomy and can act on their own to achieve specific goals.” When considered in the context of data and analytics, AI agents enable you to ask questions of the data that you could never ask before. 

With the goal of simplifying life, AI agents blend hype and potential. By embracing the evolution of large language models (LLMs) along with machine learning and process automation, AI agents are making decisions and taking actions to help organizations achieve their goals. But it’s important to remember that AI is working with us – not in place of us – which is why human-first AI is critical for the future of AI. It’s about thoughtfully reducing rote tasks with automation, not fully replacing human intelligence. In the words of one speaker, it’s “protecting our periphery” against increasing volumes of alerts that diminish our capacity to be productive. 

Technology Investments Need a Direct Tie to Business Value

While tying your technology investments to business value may seem obvious, too many times, organizations focus purely on technical capabilities and the specific challenges they solve, such as reducing duplicate data or improving overall data quality. The problem with this approach is that you are failing to articulate the value the business receives, which, in turn, makes it difficult to gain support and buy-in for the technology solution you are pitching. Decision-makers view your proposal as a widget, not a strategic solution that will advance your data and analytics strategy. 

Instead, it is important to frame your technology request in the context of challenges the business can’t solve. For example, let’s say that incomplete data spread across disparate data silos is preventing your organization from delivering a top-notch customer experience. If this is true, lead with your inability to delight your customers and position the technology as the solution to that problem. Of course, you need to integrate your data and improve its quality, but leading with that point isn’t as compelling. 

“Boil the Ocean” is Not a Viable Strategy (as if it Ever was!)

As data leaders look to tap into the myriad of ways AI can transform their organizations, one thing is becoming very clear: trying to do everything, all at once, is impossible. Rather, organizations need to take the “start small, prove value, and grow” approach, especially when it comes to implementing AI. 

When you structure your AI adoption into manageable phases, you can introduce the business into the process, involving them in the testing of applications and refinement of results. A focused project also makes it easier to identify and resolve potential challenges, fine-tune models, and ensure alignment with business goals before a full-scale rollout. Starting with a refined scope also fosters stronger user adoption, as teams can adapt to AI-driven workflows at a pace that feels comfortable for them, building confidence and laying the groundwork for sustainable, scalable AI that drives real business impact. 

AI Literacy is Mandatory, Not Nice-to-Have

Data literacy isn’t new. But its relative, AI literacy, is gaining ground. And that’s because understanding AI’s capabilities, limitations, and ethical implications is critical as AI technologies continue to take hold in businesses worldwide. AI literacy helps organizations to manage expectations when it comes to what’s possible – and what’s not – as it relates to AI. It empowers staff with the knowledge to use AI responsibly so they can infuse it appropriately within business workflows and processes. 

Just as data literacy has become essential for making informed business decisions, AI literacy is becoming equally important for modern businesses. Understanding how AI works, its limitations, its regulatory and ethical implications, and its impact on the modern workforce is critical. By implementing AI literacy programs, businesses can empower their teams with the knowledge to interpret AI-driven insights, collaborate with AI-powered tools like AI agents, and make strategic choices that maximize value while minimizing risk, all while highlighting the risks of misuse and misinterpretation of AI-powered insights. 

Trustworthy Data is Critical, but Difficult to Obtain

Building trust in the data was another widely discussed topic at this year’s event. While everyone agrees that trustworthy data is critical (if not essential), now more than ever, there is also wide recognition that it’s difficult to obtain. For too long, organizations have skipped over the hard work of assessing, improving, and reviewing their data, jumping ahead to operationalizing it. But as many organizations have learned, this is a mistake. And it may be time to reevaluate their master data strategy. 

While it’s true that governing all of your data before using it in AI isn’t entirely realistic, there are steps organizations can take to significantly improve its quality before using it to power AI applications. At Tamr, we call this the MDM journey. And like trustworthy data, we believe it’s non-negotiable when it comes to improving your master data management (MDM) strategy and establishing trust.

Clearly, the data and analytics industry is evolving rapidly. And conversations like the ones that happened at the recent Gartner Data & Analytics Summit will shape the future. From AI agents to trustworthy data (and everything in between!), these insights reinforce the need for businesses to prioritize innovation, collaboration, and adaptability.

‍