The near future of analytics in the AI era

According to Gartner, 75% of new analytics content will be contextualised for intelligent applications through generative AI by 2027, enabling composable connections between insights and actions.

During the Gartner Data & Analytics Summit in Sydney this week, Georgia O'Callaghan, Director Analyst at Gartner, said: “We’re moving from an era where analytic tools help business people make decisions, to a future where GenAI-powered analytics becomes perceptive and adaptive. This will enable dynamic and autonomous decisions that have the potential to transform enterprise and consumer software, business processes and models.”

A Gartner survey of 403 analytics or AI leaders, conducted between October and December 2024, revealed over 50% report their organisations use AI tools for automated insights, and natural language queries for analytics or AI development. Even with these capabilities, the static nature of current analytics often falls short of delivering in a truly dynamic and automated fashion.

Gartner predicts augmented analytics capabilities will evolve into autonomous analytics platforms by 2027, which will fully manage and execute 20% of business processes. The perceptive future of analytics will deliver benefits by being proactive, collaborative, connected, contextual and continuous.

“Perceptive analytics will use AI agents and other GenAI-fuelled technologies to continuously monitor evolving conditions and perceive the target environment, such as market shifts, customer behaviour changes or supply chain disruptions,” O’Callaghan said. “Guidance and analysis can then be autonomously adjusted in response, creating a more resilient and responsive analytical infrastructure. As these capabilities emerge and are adopted by organisations, their potential to reshape business operations and drive growth will only continue to expand.”

The overarching risk of perceptive analytics

According to Gartner research, the overarching risk that applies to perceptive analytics is the over-reliance on autonomous actions without sufficient validation, which could result in unintended negative consequences, reputational damage and regulatory scrutiny.

The risk of ‘agent drift’ is a serious concern, where a system’s perceptions and actions gradually deviate from desired outcomes due to evolving data or unforeseen interactions. Guardian agents are emerging to deal with this inherent issue in AI systems, according to Gartner. These agents are specifically tasked with monitoring and enforcing policies and rules to ensure the systems operate within a set of guardrails.

“Building guardian agents will need to be a key focal point of new governance initiatives for data and analytics leaders, as agentic and perceptive analytics become the standard way of insight delivery across platforms,” O’Callaghan said.

Image credit: iStock.com/Just_Super