As the deployment of generative AI and agentic AI prompts enterprises to continuously strengthen their governance systems, companies are accelerating efforts in data management and accessibility. CarMax, an automotive retail and financial services provider, began a comprehensive overhaul of its data practices nearly a year ago when it launched a broader modernization initiative.

According to Abhi Bhatt, CarMax's vice president of technology for data and AI, the company had long relied on an on-premises data warehouse platform that faced scalability bottlenecks, leading to project delays and other concerns. Bhatt manages three teams, including one focused on customer experience.

"When you try to modernize, it's not a single component: it's the entire ecosystem," Bhatt said in an interview with CIO Dive. "It's not just the data platform; it involves the data integration platform, consumption platforms, analytics and reporting... That's why it evolved into a comprehensive modernization project."

Abhi Bhatt, VP of technology – data and AI at CarMax
Abhi Bhatt, vice president of technology for data and AI at CarMax
Image courtesy of CarMax

During the planning phase, company leadership reached consensus around three key goals: improving data accessibility, enhancing system uptime, and optimizing costs where feasible.

"The importance of data has always been there," Bhatt said. "With developments in AI, people's way of thinking has shifted—we no longer only need to consider structured data."

Bhatt noted that the shift in management's mindset was relatively easy to achieve, feeling like a natural extension of the company's existing priorities. However, actually implementing technology that can support better data processes proved more challenging.

Bhatt's team wanted to improve the quality of access and data quality for information stored in SharePoint, Word documents, and policy files.

"All this data has to be clean," Bhatt said. "It has to follow the same governance principles, including data stewardship, retention, and monitoring, so that we can effectively use generative AI."

Companies across industries want to avoid the negative impacts of poor data management. According to a survey by Semarchy of enterprise decision-makers, persistent data issues are leading to declining trust in AI outputs, project delays, and rising costs. Due to the urgent need for remediation, fewer than half of the business leaders surveyed said they can meet their AI goals this year.

CarMax chose to migrate its data pipelines and business intelligence reporting to the cloud with the help of Snowflake. Snowflake subject matter experts helped train engineers and end users on effective tool usage. Additionally, the company brought in a consulting partner to support the engineering team and ease the pressure of change management.

After the migration, CarMax continued to hold training sessions as needed and designated "advocates" within different business areas to educate colleagues, further reducing adoption concerns.

"Once you have the right platform and the data is in place, we can more easily start thinking about data quality, data governance, and the next steps in this evolution," Bhatt said.

Exploring AI

In recent months, several automotive companies have accelerated their AI initiatives amid tariff-driven economic uncertainty. Automakers such as Ford, General Motors, and Toyota are exploring use cases with the potential to optimize costs and enhance customer experience.

Peers like online car retailer Carvana are also actively embracing the technology.

Although CarMax has deployed traditional AI and machine learning technologies over the past decade, the recent data modernization has paved the way for generative AI adoption.

"CarMax is applying generative AI across multiple use cases at different levels of maturity," Bhatt said.

Two key initiatives are enhancing access to knowledge bases and boosting productivity through Microsoft's 365 Copilot and AI coding assistants.

Bhatt noted that platform fragmentation can be a potential disruptor to data accessibility and management. The company is working to address this by consolidating onto as few platforms as possible. Additionally, CarMax plans to refine its governance practices around unstructured data to mitigate data stewardship challenges.

"Before I end my day, I emphasize the importance of data and data quality for AI," Bhatt said.