(This is a three-part series on Decision Intelligence.)

After years of sustained investment in data and analytics, most organisations have built mature capabilities: robust data/BI platforms, advanced analytics engines, compelling visualisations, and increasingly advanced models. Yet, despite this progress, many still struggle to make decisions that are consistently better and faster. The challenge is not the capability itself, but how analytics is positioned and integrated within the organisation.

The reality is that most analytics teams are designed to optimise the flow of information, not the quality or speed of decisions. They produce insights, reports, and forecasts, but often operate outside the decision-making moment. By the time insights are delivered, the opportunity to influence outcomes has either passed or becomes diluted across layers of governance and competing priorities.
This gap between insight and action – what we might call decision latency – is both measurable and material.

This gap between insight and action – what we might call decision latency – is both measurable and material.

Research suggests that operational decisions can take up to two days, tactical decisions around a week, and strategic decisions close to three weeks in many organisations (Cognizant/Oxford Economics, 2023). At the same time, only 37% of organisations report making decisions that are both fast and high-quality, highlighting speed as a systemic constraint rather than a data problem (McKinsey & Company, 2019).

The cost of this latency is significant. According to IBM research, 80% of organisations rely on stale data at least some of the time, and 85% of data leaders say delays in data availability have directly impacted business outcomes (IBM, 2023). Other studies estimate that slow decision-making can cost organisations up to 5% of annual revenue.

This is not an analytics failure. It is a design failure.

Decision Intelligence requires a shift from producing insights to embedding intelligence directly into how decisions are made. It reframes analytics from a support function into a core component of the decision system – integrated with strategy, ownership, governance, and execution.

In practice, this means moving beyond dashboards and models toward decision-centric design where analytics is embedded directly into workflows, aligned to strategic intent, and connected to clear decision rights and accountability. This is also where Agentic AI comes in, enabling decisions to be augmented, orchestrated, and in some cases autonomously executed within defined governance and control frameworks.

It also requires a fundamental shift in how success is measured. Rather than focusing on outputs such as reports delivered, dashboards published, or models deployed, organisations must measure impact at the level of decisions, including speed, quality, consistency, and outcomes.

The question for leaders is no longer whether their organisation has strong analytics capability. It is whether that capability is structured to influence decisions at the moment they matter. Because without that shift, even the most advanced analytics will continue to optimise information delivery while decisions remain slow, fragmented, and inconsistent.

References

• Cognizant & Oxford Economics (2023), The Decision-Making Disconnect
• McKinsey & Company (2019), Decision Making in the Age of Urgency
• IBM (2023), The Cost of Delayed Data