Process mining in practice: what it finds that Six Sigma misses.

This is where the distinction between Lean Six Sigma and process mining becomes structurally important.

Lean Six Sigma has long established itself as a rigorous and effective methodology. It thrives in environments where the problem is already framed — where there is a known inefficiency, a measurable deviation, or a performance gap that can be isolated. Its strength lies in discipline: defining, measuring, analyzing, improving, and controlling. It brings structure, statistical robustness, and a proven ability to deliver sustainable gains.

But it carries an implicit assumption: that the problem is already visible.

Process mining challenges that assumption. By reconstructing processes from system-generated event logs, it does not start from what people think is happening, but from what is actually happening. And that distinction is not trivial. In many organizations, the gap between the documented process and the real process is not marginal — it is structural.

What process mining reveals is often uncomfortable. Variants multiply. Exceptions become the norm. Bottlenecks appear in places no workshop had identified. Rework loops, delays, and hidden dependencies emerge not as isolated issues, but as systemic patterns. It is not that organizations are unaware of inefficiencies — it is that they underestimate their scale, frequency, and interconnectedness.

This is precisely why process mining often creates an initial shock of clarity. It does not optimize the process; it reframes reality.

And yet, this is also where many initiatives begin to fail.

Because visibility, in itself, does not transform anything.

A recurring pattern across industries is that companies invest in tools — sometimes advanced platforms such as Celonis or SAP Signavio — and quickly achieve a level of transparency they never had before. Dashboards are built. Bottlenecks are quantified. Deviations are mapped.

But then the dynamic slows down.

The reason is not technical. It is organizational.

Process mining surfaces issues that cut across silos. It exposes inefficiencies that are not owned by a single team. It challenges existing narratives about performance. And most importantly, it requires decisions that often involve trade-offs — between speed and control, cost and quality, local optimization and global efficiency.

Without a governance model capable of absorbing these tensions, insights remain just that: insights.

This is where Lean Six Sigma re-enters the picture, not as a competing approach, but as a necessary counterpart. If process mining is about discovering where value is lost, Lean Six Sigma is about structuring how value is recovered. It brings the discipline required to turn observations into actions, and actions into sustained results.

However, even this combination is not sufficient on its own.

Organizations do not fail in these initiatives because the methodologies are flawed. They fail because they underestimate the conditions required to make them work.

Data, for instance, is often assumed to be readily usable. In practice, extracting meaningful event logs requires a deep understanding of systems, data models, and business semantics. What appears as a technical prerequisite quickly becomes a transformation effort in itself.

Then comes the question of ownership. When a process spans multiple departments, who is accountable for improving it? Without a clear answer, improvement initiatives dissolve into coordination overhead.

And finally, there is resistance — not explicit, but structural. Process mining does not just highlight inefficiencies; it redistributes visibility. It makes certain dysfunctions undeniable. That alone can be enough to trigger inertia.

This explains why many organizations turn to external support. Not because they lack internal capabilities, but because they need a mechanism to align perspectives, accelerate decision-making, and bridge the gap between data and operations. Consultants, in this context, are less problem-solvers than integrators — connecting analytical insight with operational execution and governance.

Still, even with the right support, a fundamental risk remains: confusing analysis with transformation.

One of the most common failure modes is what could be described as “analytical saturation.” The organization becomes highly capable of diagnosing issues, but progressively less effective at resolving them. Dashboards multiply, but decisions do not. Insights become more precise, but impact stagnates.

At that point, process mining turns into a reporting layer rather than a transformation lever.

Avoiding this trap requires a shift in perspective. Process mining should not be positioned as a tool, but as an entry point into a broader transformation logic. Its role is to challenge assumptions, to surface hidden complexity, and to create a fact-based foundation for decision-making. But the transformation itself happens elsewhere — in governance, in operating models, in execution discipline.

Lean Six Sigma, in contrast, should not be seen as an alternative, but as the mechanism that ensures that what is discovered is actually addressed.

The real question, therefore, is not which methodology to use. It is how to sequence them, and how to embed them into a system that connects visibility, decision, and action.

Organizations that succeed are not those that deploy process mining the fastest, or apply Lean Six Sigma the most rigorously. They are those that understand the interplay between the two — and more importantly, the organizational conditions required to make that interplay effective.

In the end, the value of process mining is not that it finds problems. It is that it reveals how incomplete our understanding of those problems usually is.

And the value of Lean Six Sigma is not that it solves them. It is that it provides a way to do so consistently.

Between the two lies the real challenge of transformation: not seeing more, or even doing more — but aligning what is seen with what is done.