Process mining or expert interviews: when is system data enough?
System data is enough when your question is what happened, how often and how long — in a process that runs inside your systems. You need the expert when the question is why: why a case took a different path, which exception applied, what the right decision is next time. Mining shows where to look; interviews explain what you find there. In SMEs, with lower volumes and much of the work outside systems, the interviews usually carry more of the answer.
What system data actually delivers
Process mining and simpler analyses of system data give facts no interview can: how many cases took which path, how long each step really took, where cases loop back, and which customers or products behave differently. They are free of memory bias and self-image, and they cover every recorded case.
They cannot explain a single one of those facts. System data is enough if your question is about volumes, durations and paths, and the decisions in the process are already rule-based and recorded in the system.
What expert interviews actually delivers
An interview with the person who handles the cases gives the reasons: the rule behind a decision, the exception that sent a case another way, the history of a customer, what they would do differently next time. It also covers everything that happens outside the systems.
Interviews have their own bias: experts describe the ideal process and forget what they do automatically. The remedy is to talk about real, recent cases rather than general practice — see how to interview an expert properly.
System data vs expert interviews: at a glance
| Criterion | System data and process mining | Expert interviews |
|---|---|---|
| Answers | What happened, how often, how long | Why it happened, and what to do next time |
| Source | Event logs and records | The people who decide and do the work |
| Bias | Blind to anything not recorded | Memory and the tidy version |
| Effort | Data preparation | The expert's time |
| Work outside the systems | Not covered | Covered |
| Rare cases | Visible as outliers, unexplained | Explained, if the expert met them |
| Best combined with | Interviews on the outliers | Data to choose which cases to discuss |
What happens when you choose wrong?
A manufacturer's data showed that quality holds on one product line took far longer to clear than on the others. The obvious conclusion was a capacity problem, and a second inspector was proposed.
A conversation with the quality lead told a different story. Holds on that line waited for one engineer, the only person who could approve deviations against that customer's specification. A second inspector would have changed nothing. Capturing the engineer's approval rules, and naming a back-up, did. (An illustrative example, not a client case.)
When do you need both?
Use data to choose the cases, and interviews to explain them. Start from what the data shows — the outliers, the loops, the slow steps — and discuss exactly those cases with the people involved. Then check what the experts say against the data. Each source keeps the other honest. Where the two disagree, you have usually found the most valuable knowledge in the process.
Where MentX fits
MentX is a personality-aware knowledge base. It captures the knowledge of a company's key person — from meetings and a guided intake track — into a living, temporal knowledge graph with full source attribution.
Conversations and meetings are where the capture happens — see how it works. System data helps choose which cases to discuss, and every captured statement keeps its source, so you can see what came from the data and what came from the expert.
Questions people ask
How do we keep an interview from describing the ideal process?
Walk through a real, recent case step by step, and ask about the last exception rather than the usual way. People describe what they did far more accurately than what they generally do.
Is process mining affordable for a small company?
Tools exist for smaller companies. The larger cost is usually preparing the data and having enough cases in one system to show patterns — which is why many SMEs get more from interviews first.
Who should we interview?
The people who handle the cases, not only their managers: the planner, the buyer, the technician, the assistant who checks every export order. Add the person who decides on exceptions, and anyone who is asked for help more often than others.
