The 2025 Gartner CDAO Agenda Survey found only 55% of data and analytics teams rate governance effective, while building analytics scored 85%, the lowest and highest scores across 14 measured capabilities respectively, according to reporting by IIoT World. At the Gartner Data & Analytics Summit in Orlando, Gartner analyst Stephen Kennedy explained that many data governance programs stall within nine to 12 months because they default to command-and-control structures that the broader organization resists. Kennedy advised reframing governance away from abstract policy exercises and toward specific business outcomes; he is quoted saying, "The first rule of data governance is to not talk about data governance." The article contrasts that approach with AI governance, which the piece frames as requiring a different organizational structure, per Kennedy's remarks at the summit.
What happened
The 2025 Gartner CDAO Agenda Survey found only 55% of data and analytics teams rate governance effective, while building analytics scored 85%, IIoT World reports. At the Gartner Data & Analytics Summit in Orlando, Gartner analyst Stephen Kennedy explained that many data governance programs are deprioritized within nine to 12 months after launch because they rely on centralized, command-and-control structures that face organizational resistance. Kennedy is quoted saying, "The first rule of data governance is to not talk about data governance." The summit coverage describes AI governance as requiring a fundamentally different organizational structure than traditional data governance, per Kennedy's explanation.
Editorial analysis - technical context
The article highlights a recurrent operational failure mode in governance programs: translating policy into business-impacting data touchpoints. Industry-pattern observations note that governance initiatives framed as abstract rulebooks rarely change day-to-day data handling; instead, tying policies to specific KPIs and business processes tends to improve adoption. This framing matters for practitioners building pipelines, access controls, lineage, and instrumentation because buy-in is often driven by measurable business objectives rather than compliance rhetoric.
Industry context
Reporting from the summit frames the contrast between data governance and AI governance as organizational rather than purely technical. Industry observers have noted similar distinctions: AI governance frequently requires cross-functional risk evaluation, model lifecycle controls, and runbook integration that sit differently within product and engineering teams compared with traditional master-data or metadata governance. For practitioners, that means governance tooling, telemetry, and decision checkpoints may need different placement and ownership models to be operational.
What to watch
Follow formal guidance and playbooks published after the summit for concrete ownership models and KPI mappings. Observers should also watch updates to enterprise governance tool capabilities that explicitly map policies to business KPIs, and any case studies showing governance programs sustaining beyond the nine- to 12-month drop-off window.
Scoring Rationale #
The story highlights a notable governance capability gap revealed by Gartner survey data and summit commentary, which matters for practitioners designing controls and ownership models. It is a significant operational issue but not a frontier technical development.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.