Poor data quality will continue to be a major hurdle preventing the deployment of advanced analytics, including artificial intelligence (AI), through 2025, according to a new report by Gartner.
The research firm emphasised that for enterprises to advance their AI initiatives, data and analytics (D&A) leaders must prioritise three interdependent journeys: business outcomes, D&A capabilities, and behavioural change.
“AI continues to drive enterprise planning, with more than half of CEOs believing the technology will most significantly impact their industry over the next three years,” said Carlie Idoine, VP analyst at Gartner. “With this in mind, D&A leaders are uniquely positioned to drive maximum impact on business outcomes due to their proximity to this technology.”
Gartner highlights that demonstrating AI’s value remains a top challenge for enterprises. To address this, D&A leaders should focus on establishing trust, measuring productivity improvements, and effectively communicating the value of AI initiatives.
“Demonstrating the value of AI continues to be a top barrier to implementation,” said Idoine. “D&A leaders must focus on building the right trust levels based on context as the first step to demonstrating value.”
Poor data quality often leads to AI failures, making trust models essential in assessing data value and risk, providing ratings based on lineage and curation. Leaders should also weigh AI investments against cost, complexity, and risk to assess competitive impact, while ensuring that AI-related costs, including those for data management, governance, and change management, are thoroughly considered and communicated.
Ensuring a robust and scalable AI technology stack is essential for D&A leaders. Gartner underscores the importance of adaptability and modularity in AI ecosystems.
“Stack versus best of breed is not new, but the dynamics of this decision are,” said Gareth Herschel, VP analyst at Gartner. “D&A leaders must cultivate an adaptable ecosystem that scales in order to meet the demands of creating the best AI offerings possible.”
To achieve this, organisations need to update or replace architecture components to adapt to evolving AI requirements. Integrating trust into FinOps, DataOps, and PlatformOps will help transition from a conventional tech stack to a trust-based AI stack. The use of dynamic AI agents can enable a more responsive AI ecosystem, leveraging active metadata to adapt to changes.
Beyond technology and governance, addressing the human element is crucial for AI adoption. Gartner notes that fostering AI literacy, encouraging new skill development, and promoting collaboration will be key for successful AI deployment.
“AI is transforming everything, and people are expected to transform too,” said Idoine. “But people are not the same, and we engage with data and analytics in different ways.”