From Inside-Out to Outside-In: Conceptualizing Journey-to-Process Mappings from a Process Observability Perspective
In today’s fast-moving, interconnected world, it is crucial to not only have a tight control on organizational operations, but also ensure operations can be transformed and continuously improved to serve the hard objectives, subjective journeys, and experiences of organizational stakeholders. Treating the latter (stakeholder objectives, journeys, and experiences) as constraints of the former (operations) reflects business process management’s traditional, efficiency- and order-oriented attitude. However, seeing the optimization operations and their processes as more than isolated objectives is crucial to unlocking long-term and sustainable business value. After all, organizational processes are primarily abstract entities supporting the objectives and affecting people’s experiences, be it customers, employees, or other stakeholders, that define the organization’s success. To better integrate journey-oriented and process-oriented perspectives, we need to map journey data and knowledge to process data (and models). Below, we describe what exactly can be mapped, how the mapping process can be automated, and how the integrated data and knowledge analyses can lead to more useful insights and better process observability.
Journey-to-Process Mappings: What Can We Map?
Journey-to-process analytics fuses different data and knowledge types for the best business performance, analogously to how self-driving (or semi-autonomous) cars fuse data from different sensors with maps and other hard facts to facilitate a safe driving experience, even in challenging environments. Let us provide an overview of the mapping and fusion options and the corresponding analysis methods (see the figure below), starting with the baseline of raw eXperience (X) data.
X Data, Unfused. Journey-to-process analytics starts with experience data as the cornerstone of enabling an outside-in perspective on organizational operations. The central premise is that journey-to-process analytics is impossible without data that provides at least the basis for rough intuitions about stakeholder experiences. Here, the obvious standard analytics approaches can be applied. In the case of text data these can be, for example, sentiment analysis, topic extraction, summarization, and visualization (as word clouds). In case we have quantitative satisfaction data (metrics), benchmarks, trends and anomaly analyses can be executed, possibly segmented by journey variant. Beyond that, organizations may tap into implicit, pre-existing models of business processes or journeys that exist in a general context (e.g., in reference model collections): this can build the first bridge between experience and operations. For example, relevant reference models can be recommended that provide design blueprints for addressing some of the issues elicited from textual experience data, or general changes to processes and journeys can be suggested without relying on an underlying organization-specific model.
X Data-to-Journey Model. If one or several journey models are present, experience datapoints (or sets thereof) can be first mapped to a model and then to a particular journey step or stage (collection of steps). This can highlight where exactly the issues in a journey occur. While the operational aspects are shallow in this mapping, processes (or process steps), IT systems, and roles that the journey model references may provide a first glimpse into the outside-in view.
X Data-to-Process Model. An analogous mapping can be generated from experience data to process models and activities therein (tasks in BPMN diagrams). Here, intermediate models such as journey models or definitions of IT systems and roles with links to process models or steps thereof may be required. Afterall, it is uncommon that stakeholders specifically mention a particular end-to-end process or process step; references to IT systems and other objects like roles or attributes like average waiting times are more common. The mapping can then identify the operational source of experience issues, speeding up the search for the underlying root cause and the necessary change that fixes it.
X Data-to-Process Data
The most advanced and impact-wise most crucial- journey-process mapping is the fusion of experience and process data. In ideal cases, experience data is collected in a process-oriented manner, i.e., a case identifier (case ID) exists as an attribute to every point of experience data, which then allows for the direct augmentation of event logs and the utilization of experience data in process mining, e.g., for correlating process performance indicators (PPIs) or conformance measures and satisfaction metrics, the segmentation of process variants by experience ‘topic’, and for anomaly detection. Without case ID, experience data can still be placed in the context of process-level aggregations, for example to relate it to overall performance metrics (potentially based on benchmarks) and map sentiment or topics to nodes in the mined directly-follows graphs (DFGs).
Enhancing Process Observability by Fusing Journey and Process Data
Given the mapping options that we have laid out above, we can outline a hierarchical approach for organizations to increase their process observability, i.e., their ability to view organizational operations from multiple perspectives that are then combined into a holistic, actionable view – analogous to how self-driving cars rely on the fusion of data from different sensors. Here, we envision the following ladder of Journey-to-Process Observability (also see the figure below):
- Utilizing Experience Data with neither Models nor Process Data Analytics. For this, no process models or data are required, meaning that organizations with very low levels of process management maturity can start immediately. For meaningful operational insights, comprehensive collections of general process knowledge must be available, as provided by SAP Signavio Process Explorer.
- Utilizing Experience Data with Readily Available Data-Driven Process Insights. The latest frontier of real-world process intelligence is the provision of data-driven insights and benchmarks directly on top of state-of-the-art enterprise systems, as provided by SAP Signavio Process Insights. If such insights are available, experience data can be mapped to them, for example to compare journey and process pain points and to prioritize operational issues based on their impact on stakeholder experience.
- Utilizing Experience Data with either Modeling or Mining. If an organization maintains their own models of processes or journeys (in SAP Signavio Process Manager), or if at least some event logs are available (in SAP Signavio Process Intelligence), more in-depth and context-tailored journey-to-process insights can be generated. In the case of model availability, the insights are primarily qualitative, telling us where in an organization particular experience pain points (or highlights) are. In case event logs are available, nuanced quantification, as well as drilldowns into variants are possible.
- Utilizing Experience with both Process Models and Data. Finally, the highest level on the ladder of journey-to-process observability fuses experience process data of various granularity and provides a comprehensive inside-out map of an organization’s operations based on process and journey models that provide crucial connects between different data analysis views, enabling a holistic perspective on journey-to-process observability.
In this blog post, we have provided a motivation and an overview of different ways to fuse journey (experience) and process data and knowledge to generate more meaningful and actionable insights about business operations and customer (or more generally: stakeholder) experience alike. SAP Signavio’s unique journey to process capabilities provide some of the most crucial journey & process fusion options already – with yet more to come!