05 · Getting processes ready for AI
Process Automation
Automation amplifies whatever is already there – for better or worse. So my work does not start with the tool, it starts with the process: understand it, measure it, clean it up, standardise it. Only then can the question of AI and automation be answered sensibly.
Where I put the focus
Clarity before technology
Before any tool talk: how does the process actually run, where do exceptions arise, what is stable enough to automate?
Data quality as a precondition
Automation needs dependable data. We check measurement systems, definitions and sources before anything gets built.
Sober assessment
Not every step is worth it. We rate candidates by volume, variance and value – and write down what deliberately stays manual.
Measured, not assumed
Before and after on the same metric. What cannot be shown does not count as an improvement.
How I approach it
-
01
Understand the process
Map the current state with the people involved, surface variants and exceptions.
-
02
Check the data
What data exists, how reliable is it, what is missing for automation?
-
03
Assess candidates
Prioritise steps by effort, stability and expected effect.
-
04
Start small
One contained use case, clearly scoped, with a defined metric and an honest review.
What I bring – and what I don't
I am a process consultant, not a systems integrator. My role sits upstream of the technology: understand the process, check the data, assess automation potential and structure the work with your IT or your vendors.