The instantiation of process-based management must be an active management system.

The benefits of process-based management cannot be realized through static diagrams and theoretical concepts. Action is required. Effective action must have a foundation of sound theory, but theory alone can only amount to untested aspirations.

In this column I want to describe that necessary action through the lens of three critical verbs: prioritize, predict, and prove.

Prioritize

Any organization has hundreds of processes; indeed, we might say thousands if we dig a little deeper into the hierarchy. It makes no sense to say that they can all be actively managed. There are just too many.

The good news is that we need only manage a relatively small number of high-impact processes. Many large and complex organizations are discovering that they can identify 20–30 high-impact processes and they focus on them.

Even 20 processes are likely to be too many all at once when first starting the process-based management journey. Pick a handful to start with and once they are under active management, pick another handful.

Also worth remembering that if a processes at any level in the architecture hierarchy is performing well according to its KPIs and targets, then it is reasonable to assume that its subprocesses are performing well enough.

In deciding where to apply scarce process management, analysis, and change resources we need to prioritize. Which processes should we analyze? Which performance anomalies should we investigate? Which opportunities should we pursue?

Where can we get the best return on process?

Practical steps

• Maintain a set of selection criteria to be used in identifying a limited collection of high-impact processes.


• Develop process-based management capability incrementally, fine tuning the approach and collecting evidence of success.

Predict

The fundamental purpose of performance management is not to know what has already happened, but rather to be able to predict with acceptable certainty what will happen next. Knowing what happened yesterday is useful; knowing what will happen tomorrow is invaluable.

While it is important to understand and learn from what has already happened, an essential test of effective management is whether future performance can be predicted. If not, management must be reactive rather than active, indicating a low level of control over the process and its outputs.

If an unpredicted change occurred, what can that teach us about the process and its ecosystem? Why wasn't it predicted? Do we want to eliminate the cause of the change or amplify it? Conversely, if we predicted a change (a process improvement, perhaps), did it happen as anticipated, and if not, why not? This is much more nuanced management than the simple tracking of a set of KPI traffic lights. This is predictive process management.

A process may have performance that is predictable but unacceptable. However, if performance is not predictable the process is unstable and that will certainly never be acceptable.

So, process performance first needs to be shown to be predictable within known performance limits, and then we can make it acceptable. By what magic is this possible?

Not magic, but it feels close to it. The device that enables this is the Process Behavior Chart (PBC) — the closest you can get to magic while remaining statistically significant! I've written about this elsewhere and you can read the theory and practice of using PBCs in Signals in the Noise.

As useful as the PBC is, we need to do more than drive using the rear-view mirror. We need to be studying the map, the weather, engine performance, anticipating changes and challenges.

In a time long ago, I had a private pilot license and although I haven't flown for years the concept of being the pilot-in-command has never left me. The realization that uncontrolled, unanticipated changes can ruin your day, if not your whole life, and that only you can make a difference, brings the idea of prediction into sharp focus as you deal with moving in three dimensions, with time being critical, weather systems changing, other traffic in the airspace, air traffic control issuing requests and instructions, and fuel supplies always diminishing.

We need a lot more predictive pilot-in-command thinking in process management. Success in any process, on land or in the air, requires the identification and prediction of the intersection of the many variables that influence performance.

Practical steps

• Select a couple of demonstration processes and draw PBCs for their process KPI targets. Test for performance stability and acceptability.


• Track process performance with the PBC and refine the ability to predict future performance.


• Develop an enhanced understanding of the circumstances that influence the performance of the process.

Prove

The only acceptable measure of process management success is proven, valued, business, benefits. Equal emphasis on all four elements. Proven beyond any doubt to have been delivered. Valued by the relevant business managers, not just the BPM team. Business impacts that make a difference. Benefits to organizational performance, not just new artifacts.

In promoting the adoption of process-based management, we must keep it real. In asking the organization to reimagine how it is managed, we must be able to prove that it is worth the effort. The reality must be that there are solid business benefits that solve problems and make management easier. Process-based management must be a solution, not a new problem.

Proving the benefit must begin during the process analysis phase. In defining a problem to be solved or an opportunity to be realized there must be a clear definition of the benefits to be achieved and how that will be measured. The business case for change must show, in detail, how the change benefit will be proven.

Approval for a project to improve (change) a process is (or at least should be) about approving the delivery of the benefit, not just delivery of the project. If we can't be sure of the change benefit, don't run the change project.

If a change is predicted to create a saving in time or money, the business case should also say how that newly 'free' time or money will be invested.

Beware the business case that depends on the aggregation of small numbers. For example, a change project might deliver a time saving of two hours per month for each of 5,000 staff giving a total monthly saving of 10,000 staff-hours. This could be said to be a saving of some 60 FTE. We can keep doing the arithmetic and turn that into a budget saving.

However, what is likely to happen in the real world is that no budget will change because of the savings. The savings are real but difficult (perhaps impossible) to harvest as they are small fractions of an individual's total work hours. If the project was approved based on a promised saving of 60 FTE, then it must fail.

Of course, the benefit need not be a financial one. We might have said that we'll reinvest the two hours per month in an all-staff meeting that will have a positive impact on organizational culture — and we'd also need to forecast how that would be measured.

Benefits aren't real unless proven, and not just proven to exist in theory, but in the practical reinvestment of the delivered benefit.

Practical steps

• Before a change project is approved, demand a clear and detailed definition of how the value of the change will be measured and proven.


• Include in the business case for a process change project a clear statement of how the harvested benefit is to be used.

Building a Management System

The instantiation of process-based management must be an active management system.

Prioritize to ensure optimum return on process. Enable agile and responsive management that will not only track but predict process performance. Prove the value by delivering genuine business benefits.

# # #