Key points
- A survey report from MIT Technology Review's custom content arm, sponsored by Snowflake, finds 48% of analytics agent insights lead to business-improving actions, up from 37% a year earlier.
- A small group of 'insight leaders' sees more than 70% of agent insights lead to improvement, with the widest gap in prescriptive insights.
- Lack of business context (55%) and fragmented or inaccessible data (45%) are the top reasons for weak insights.
- 52% of organizations say agents produce too many insights to evaluate; only 24% can review all of them.
Why it matters for marketers: Marketing teams adopting agents for analysis and action need business context, governance and review capacity in place first, or they risk acting on weak recommendations.
A new report from MIT Technology Review Insights, produced in partnership with Snowflake, looks at how well analytics agents perform inside companies. On average, 48% of the insights those agents generate lead to actions that improve the business. A year ago, respondents put that figure at 37%. Progress is real, but more than half of insights still fail to help. The content was produced by the publication's custom content arm, not its editorial staff.
The report separates two kinds of output. Most respondents rate descriptive insights well, but views on prescriptive insights, the ones that drive actions, are mixed. That gap matters for anyone letting agents adjust campaigns, budgets or content. The report warns that flawed insights that trigger wrong actions can cause reputational or financial damage.
Respondents point to context as the main culprit. Some 55% of executives cite missing business context when insight quality is poor, and 45% name fragmented or inaccessible data. A small group of insight leaders, where more than 70% of insights lead to improvement, reports fewer context problems.
Volume is a second problem. At 52% of organizations, agents produce more insights than can be properly evaluated, and only 24% can review all of them. Most review case by case, which takes heavy human effort.
For next steps, 60% of surveyed companies are prioritizing stronger AI and data governance, and 45% are adding a semantic layer to their data architecture. Marketing teams planning agent rollouts could ask the same questions: what context does the agent have, and who reviews its recommendations before action is taken?
Sources
This brief is our own summary. Read the full reporting at the source.
