For more than two decades, organizations have focused on optimizing execution. We adopted Agile, embraced DevOps, automated workflows, and built increasingly sophisticated systems to help teams deliver faster and more efficiently. Yet despite all these advancements, projects still miss deadlines, budgets still slip, and strategic initiatives still fail to meet expectations. Why?
As Dejan Dimić, Engineering Manager, argued during his Tech Stories session, the problem is often not execution itself. The problem starts much earlier, in the planning phase.
Well known to many levi niners from his appearance at our 9Inspiration conference, Dejan recently returned to continue the conversation – this time exploring how AI can help organizations make better decisions before execution begins.
From Understanding the Past to Shaping the Future
The evolution of decision support has followed a clear path. Business Intelligence helped organizations understand what happened. Analytics helped explain why it happened. Predictive Analytics allowed us to estimate what might happen next.
Planning Intelligence introduces a different question altogether: Should we commit to this plan?
Rather than focusing on reporting, forecasting, or monitoring, Planning Intelligence seeks to evaluate the quality of decisions before execution begins. The goal is not to create another dashboard. The goal is to help leaders make better-informed decisions by exposing risks, dependencies, assumptions, and readiness gaps that may otherwise remain hidden.
Why Great Teams Still Miss Their Goals
During the session, Dejan presented a scenario involving the launch of an AI assistant. At first glance, the project appeared healthy. Multiple teams were aligned around a single deadline, responsibilities were assigned, and progress indicators suggested everything was moving in the right direction.
A closer look revealed a different reality.
A single architect was required across several teams simultaneously. Critical infrastructure would not be available when development is needed. Security reviews depended on vendor processes that had not yet started. Data migration activities were scheduled after integrations that depended on the migrated data. None of these issues were hidden. The information existed, but it was scattered across project plans, documentation, tickets, and institutional knowledge.
The challenge was not a lack of information. The challenge was the inability to connect the dots before execution began.
AI as a Decision-Support Partner
This is where AI can create significant value. The discussion emphasized a perspective that goes beyond content generation, automation, or coding assistance. AI’s role is not simply to help teams work faster. Instead, it can help companies understand the implications of their choices before they commit to them.
By analyzing information distributed across multiple systems and sources, AI can identify hidden dependencies, reveal resource bottlenecks, validate assumptions, and highlight organizational risks. It can provide recommendations that might otherwise require weeks of manual analysis.
Importantly, Planning Intelligence does not advocate replacing human judgment.One of the central messages of the session was that AI should support decision-making, not own it.
The analogy Dejan used was GPS navigation. A navigation system can recommend a faster route, warn about traffic, and estimate arrival times. However, the driver still decides whether to follow the recommendation. In the same way, AI can surface risks and suggest options, but leaders remain responsible for the final decision.
Measuring the Quality of a Plan
One of the more thought-provoking ideas discussed was the lack of planning metrics within most organizations. Teams spend considerable effort measuring delivery performance, productivity, and outcomes. Yet few organizations systematically measure whether a plan is fundamentally sound before execution starts.
To address this gap, Dejan introduced several concepts designed to evaluate planning quality:
- Planning Risk Index evaluates the likelihood that a project will experience planning-related issues.
- Dependency Complexity measures how vulnerable a plan is to cascade delays caused by interconnected teams and activities.
- Organizational Readiness assesses whether the organization can realistically begin execution immediately.
- Assumption Confidence measures how many critical assumptions have been validated versus accepted without evidence.
- Execution Probability estimates the likelihood of success and identifies actions that could improve the outcome.
These metrics are not intended to replace leadership decisions. Instead, they provide leaders with better information so they can make those decisions with greater confidence.
A New Governance Discipline
Perhaps the most compelling takeaway from the session was that Planning Intelligence should not be viewed merely as another AI use case. Instead, it represents a potential shift in how organizations govern change, investments, and strategic initiatives.
Just as Business Intelligence transformed the way organizations understand their past, Planning Intelligence has the potential to transform how organizations decide their future.
As AI capabilities continue to evolve, the greatest opportunity may not lie in automating more tasks or generating more content. The real opportunity could be helping organizations make better decisions before committing resources, assigning teams, and launching initiatives. Perhaps the future of AI is not about doing more work for us, but about helping us ask better questions before the work begins.
Many thanks to Dejan Dimić for sharing a thought-provoking perspective that challenged us to look beyond execution and rethink how decisions are made in an increasingly AI-driven world.
In this article:
