Agentic analytics insights fail to improve business half the time
A new report reveals that while agentic AI is advancing fast, fewer than half of its generated insights currently lead to tangible business improvements, highlighting critical gaps in context and governance.
Generative AI has already reshaped enterprise analytics by making data accessible to a wider audience. Agentic AI promises to take this further by automating workflows and executing actions based on analytical insights. However, the quality of these automated outputs remains a significant hurdle for organizations hoping to maximize their investment.
According to a recent report, less than half of the insights currently produced by analytics agents result in business improvement. While respondents generally praise the descriptive capabilities of their agents, reviews of prescriptive insights—those guiding specific actions—are mixed at best. Flawed recommendations that trigger incorrect decisions can lead to costly reputational and financial damage for companies relying on these tools.
Agentic analytics is only just starting to deliver
The current state of agentic analytics shows promising but incomplete progress. On average, 48% of insights generated by analytics agents at surveyed companies now lead to actions that improve the business. This represents a marked advance from the 37% average recorded one year ago. Despite this growth, more than half of the insights produced today are still failing to help businesses achieve their goals.
Insight leaders outperform the average
A small cohort of organizations identified as "insight leaders" demonstrates significantly higher performance levels compared to the broader market. In this group, more than 70% of agent insights lead to actions that improve business outcomes. These leaders also rate the overall quality of their agents' insights much higher than total respondents do. The disparity is especially pronounced regarding prescriptive insights, where the quality gap is widest.
Context and data access are primary bottlenecks
When analytics agents produce substandard insights, a lack of business context is the leading cause, cited by 55% of surveyed executives. Fragmented or inaccessible data is identified as the second major factor, according to 45%. Notably, fewer insight leaders report that their agents struggle with access to necessary context compared to other organizations.
Overload and governance remain key challenges
Insight overload further inhibits agent effectiveness across a majority of surveyed organizations. At 52% of companies, analytics agents generate too many insights to be properly evaluated and acted upon. Less than one-quarter (24%) are able to evaluate all agent insights; most do so on a case-by-case basis using a process that is currently human-intensive.
To address these issues, surveyed companies are prioritizing initiatives focused on strengthening AI and data governance, cited by 60%. They are also adding a semantic layer to their data architecture, with 45% adopting this approach. Experts interviewed for the report agree that implementing a semantic layer is crucial to ensuring high-performing agents in the future.
