What is process mining?
Process mining is a technique that reconstructs how a business process actually runs from the event logs kept by enterprise systems. Enterprise resource planning (ERP), customer relationship management (CRM), and ticketing systems record each step as an event. Each event holds a case identifier such as an order number, the activity, and a timestamp. Process mining software assembles these events into a map of every path that cases follow, how often each occurs, and how long cases wait between steps. Comparing the map with the documented procedure is called conformance checking.
Process mining in practice
A building supply distributor documents its order-to-cash process as six steps. A map built from a year of ERP events shows two problems the procedure never mentions. Orders from long-standing customers sit for days in a manual credit queue. Many invoices are reissued after shipment because contract prices in the system are wrong. The fixes are a credit rule for repeat customers and a pricing cleanup. Only then does the team ask which remaining steps suit robotic process automation or an AI agent. The same map informs both process optimization and use case prioritization.
What to watch for
- Work done in email, spreadsheets, or phone calls leaves no event log, so the map can look cleaner than the real process.
- Linking one case across systems, such as an ERP order and a CRM dispute, is usually the hardest data task.
- Event logs show who performed each step, so check employee monitoring rules before reporting on individuals.