Generative AI has significantly advanced enterprise analytics, making data more accessible across organizations. However, the true potential of agentic AI, which automates entire workflows and acts on insights, is yet to be fully realized. Many agent-generated insights currently fall short of driving substantial business improvements. This shortfall stems from various challenges, including insufficient business context, fragmented data, and an overload of insights, hindering effective evaluation and action. Organizations are actively seeking ways to enhance the quality and impact of these insights to unlock the full value of their AI investments.
Addressing these issues is crucial for avoiding financial and reputational damages caused by flawed insights. This report delves into how companies perceive the quality of outputs from their analytics agents and the strategies they employ to boost insight quality, ensuring their AI agents contribute more effectively to business objectives.
The Current Landscape of Agentic AI Insights
Agentic analytics is still in its early stages of delivering its full promise to businesses. While there has been a notable improvement from previous years, with nearly half of agent-generated insights leading to positive business actions, this also implies that a substantial portion—over 50%—still fails to yield tangible benefits. This indicates a significant opportunity for growth and refinement within the field. The discrepancy between the potential and current performance highlights the need for a deeper understanding of the factors that contribute to insight quality.
Furthermore, a select group of organizations, dubbed "insight leaders," demonstrate significantly higher success rates, with over 70% of their agent insights driving business improvements. These leaders also report superior quality in their agents' insights, particularly for prescriptive insights that directly inform actions. This gap underscores that while the technology has broad adoption, its effective application varies, with leaders leveraging better strategies to extract greater value.
Strategies for Improving Insight Quality
A primary hurdle in achieving high-quality insights from analytics agents is the lack of adequate business context. Over half of the surveyed executives cite this as the main reason for suboptimal insights. Complementing this, fragmented or inaccessible data presents another significant challenge, affecting nearly half of the organizations. Notably, insight leaders tend to experience fewer issues with contextual access, suggesting that a holistic approach to data integration and contextualization is a critical differentiator.
To combat these issues, companies are prioritizing initiatives focused on strengthening AI and data governance, recognized by 60% of respondents as a key area. Another vital strategy is the implementation of a semantic layer within their data architecture, chosen by 45% of organizations. Experts concur that this semantic layer is essential for fostering high-performing agents capable of delivering accurate and actionable insights. These targeted efforts aim to create a more robust and coherent data environment, enabling agents to operate with greater precision and deliver more impactful business outcomes.