Raising the quality of insights in agentic analytics
In just a few years, generative AI has changed the face of enterprise analytics, notably by making analytics accessible to users across the organization. Agentic AI promises to do even more, by automating entire workflows and taking actions based on agent insights. But there is a problem: The quality of insights that agents currently generate…
Generative AI has revolutionized enterprise analytics in recent years, making it accessible to users throughout the organization. Agentic AI, which builds upon this foundation, promises to take analytics further by automating workflows and taking actions based on insights generated by AI agents. However, the quality of these insights is a concern, as fewer than half of the agent-generated insights lead to business improvement at organizations surveyed for this report.
While most respondents praise the quality of descriptive insights, reviews of prescriptive insights—those guiding actions—are mixed. Generating business-improving insights is crucial, as failing to do so deprives organizations of their desired return on investment and can lead to reputational or financial damage. This report examines how companies perceive the quality of insights produced by their analytics agents and explores strategies to improve insight quality and enhance agent performance.
Currently, only about half of the insights generated by analytics agents contribute to positive business outcomes, a slight improvement from the previous year's average of 37%. Yet, more than half of the insights still fail to aid businesses. Organizations that can produce high-quality insights are dubbed "insight leaders." These organizations see a much higher success rate, with over 70% of agent insights driving business improvement.
The quality of insights, particularly prescriptive ones, tends to be higher at insight leader companies.
One of the primary reasons for low-quality insights is a lack of business context, cited by 55% of surveyed executives. Another significant factor is the fragmentation or unavailability of data, which affects agents' ability to provide useful insights. Insight overload, with agents generating too many insights to evaluate effectively, is also a challenge, with only a quarter of surveyed organizations able to assess all insights.
To address these issues, companies are focusing on improving AI and data governance, with 60% prioritizing these initiatives. Additionally, adding a semantic layer to their data architecture is a strategy adopted by 45% of surveyed organizations. Experts agree that a semantic layer is essential for ensuring high-performing agents. Overall, the insights from this report highlight the need for better insight quality to maximize the potential of agentic analytics.
Written by urgent.news from MIT Technology Review's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
