Introduction
A key component to a successful process mining project is having historical event log data about the process that process mining can use. The event log data is telemetry about each step that you complete in the process. The more granular the track of the steps is, the more visibility process mining analysis efforts have into how the process is working. For example, a process has Invoice Approval as a recorded step in the event log. In reality, this approval step might have 5-10 smaller steps that make up the actual approval process. In this case, process mining can't pinpoint anything inside of the larger approval process that might be the bottleneck. When you're identifying data to ingest into process mining, having access to the more granular data helps you conduct deeper analysis by using the process mining tools.
In a simple process, this data might come from a single source, such as your customer relationship management (CRM) system, which tracks each activity in a sale or a service process. In more complex processes, the event log data might span multiple systems of record. For example, in a sales process, the CRM system might track the interaction during presales, and an internal system might track the fulfillment of the order to the customer. In this example, you would need to identify and obtain event log data from both systems to have the complete process available for process mining analysis. In many cases, you might find that one or more of the systems involved is a partner or external application, such as SAP, where you need to work within the constraints of what it can provide for event log data.
Ultimately, you need to consolidate the event log data into a single stream of data that you can ingest into the Microsoft Power Automate Process Mining add-on for analysis. For lower volumes of data, you can perform this consolidation in Power Automate Process Mining by using Power Query capabilities. For higher volume or more complex data sources, you might want to do some preprocessing of the data before ingesting it into Power Automate Process Mining. For example, you might obtain data from 10 different systems of records and consolidate it into a single Azure Data Lake that the system ingests into Power Automate process mining.
As you work to identify the data to extract and use as part of the event log, it's important that you work with the broader process mining team. As a team, you should identify the data sources and what data needs to be collected. In some cases, the system of records might not collect adequate data, and it might require more enhancements before it can produce adequate data.
The following video demonstrates how to include data from extra sources. In this video, the data is from SAP.
The rest of this module explores considerations of and how to prepare your event log data for use with Power Automate Process Mining.