Predictive analytics is fundamentally about making the right predictions about future project schedules and costs which identify early warning signs, help us reduce risk, and fundamentally drive more successful project outcomes, more benefits, and a higher return on investment to the organization.
In many ways, predictive analytics is traditional risk management running on steroids. The fundamentals are the same; it is about getting the right data, at the right time in front of the right people, which is fundamental to risk management, which is fundamental to strategic planning and strategic execution.
Predictive analytics is also about being able to answer the difficult questions regarding scenario planning, risk responses, and to do that we need to harness historical data, but also harness new and emerging technologies in PPM, which is artificial intelligence, equally harness our senior leadership teams and together those three sources give us the ability to do better predictive analytics and fundamentally predict more successful project outcomes.
The importance of quality data
However, the reality is that for many organizations their existing data is not sufficient to answer those difficult questions. So the key to meeting this challenge is to identify appropriate sources or systems – what I often refer to as knowledge pockets within the organization, within the PMO, and within your projects themselves.
We are all aware of the concept of “garbage in – garbage out” and how that manifests itself in the enterprise is that we are dealing with unstructured data, which might be in the form of emails, spreadsheets, duplicate records, different systems with conflicting data – fundamentally no single source of truth.
In the PPM space we are seeing many of the large consulting houses emerged with a new concept of what we call “Predictive Project Analytics” (PPA), and fundamentally what they’re trying to do is to bridge the gap between the business need and a project outcome between the strategic planning and the strategic execution.