> ## Documentation Index > Fetch the complete documentation index at: https://docs.germainux.com/llms.txt > Use this file to discover all available pages before exploring further. # Business Process Monitoring for SAP (Configure) > Configure GermainUX to measure the execution, duration, completion, abandonment, and efficiency of SAP business processes. Components [`Engine`](https://docs.germainux.com/main/germain-ux-engine) [`JS Profiler`](https://docs.germainux.com/main/germain-ux-js-profiler) [`Mobile App`](https://docs.germainux.com/main/germain-ux-mobile-app) [`RPA Bot Recorder`](https://docs.germainux.com/main/germain-ux-rpa-bot-recorder) [`RUM JS`](https://docs.germainux.com/main/germain-ux-rum-js) Availability 2022.1 or later > **[SAP Business Process Monitoring](#configure-business-process-monitoring-for-sap)** > > Track completion and abandonment > **[Configure correlation](#configure-correlation)** > > Link events to process instances > **[Configure abandonment](#configure-abandonment)** > > Define when processes are abandoned > **[Analyze productivity](#analyze-productivity)** > > Productivity from observed interactions ## Configure Business-Process Monitoring for SAP Configure GermainUX to measure the execution, duration, completion, abandonment, and efficiency of SAP business processes. Business processes are constructed from events collected through RUM JS, the GermainUX Engine, synthetic scenarios, or custom integrations. GermainUX can analyze only the workflow steps represented by the available telemetry. ### Prerequisites Before configuring a business process: * Deploy RUM JS and/or the GermainUX Engine. * Confirm that the required SAP events appear in GermainUX. * Identify the process start and completion conditions. * Identify required, optional, repeated, and alternative steps. * Select a reliable process-correlation method. * Define the expected process and step durations. * Establish privacy requirements for business identifiers. ### Event sources | Source | Example process evidence | | - | - | | RUM JS | Pages, application views, clicks, form submissions, errors, and browser requests | | Engine | HTTP checks, API events, log events, file events, database results, and integration activity | | Custom event | Store-specific or SAP-specific business milestone | | Synthetic scenario | Controlled execution and validation of a predefined process | | SAP Commerce Cloud storefront | Product, search, cart, checkout, and order-related frontend events | GermainUX does not automatically observe undocumented SAP internals. Backend process steps require an accessible and configured source. ## Select the process Start with a process that has clear business value and observable steps. ### SAP business-process examples Potential processes include: * Purchase requisition creation * Purchase-order approval * Invoice approval * Sales-order entry * Material lookup * Inventory transfer > **Show all 11** > > * Expense submission > * Employee self-service > * Vendor onboarding > * Customer maintenance > * Service-request handling ### SAP Commerce Cloud examples For SAP Commerce Cloud, formerly SAP Hybris, potential processes include: * Product discovery * Search to product * Product to cart * Cart to checkout * Checkout to order * Customer registration * Login and password recovery * Return request The exact process must reflect the application’s actual event flow. ## Define the business process Go to: **Germain Workspace > Analytics > Business Processes** Create or select the SAP business process. Configure: * Name * Description * Application * Environment * Start condition * Completion condition > **Show all 12** > > * Failure conditions > * Abandonment timeout > * Correlation > * Expected duration > * Steps > * Relevant dimensions Use a clear name such as: ```text theme={null} SAP Purchase Requisition SAP Invoice Approval SAP Sales Order Entry SAP Commerce Checkout ``` ## Define the start and completion ### Start condition The start condition identifies when a process instance begins. Examples include: * User opens the creation page * User selects a Fiori application * User clicks Create * Relevant API request begins * A configured log event appears * Customer starts checkout Avoid starting the process on a generic login or page view unless it reliably represents process intent. ### Completion condition The completion condition identifies successful business completion. Examples include: * Confirmation page appears * Success event is created * Approved API response is received * Expected log event appears * Document status reaches the required state * Order-completion event is collected A successful click or HTTP status does not necessarily prove that the business process completed. Use the strongest available business-success evidence. ## Configure the process steps Define every observable step required for analysis. Example: | Order | Step | | - | - | | 1 | Open Purchase Requisition Application | | 2 | Enter Requisition Details | | 3 | Add Line Item | | 4 | Validate Requisition | | 5 | Submit Requisition | | 6 | Requisition Confirmation | For each step, configure: * Step name * Entry condition * Completion condition * Expected duration * Warning duration * Critical duration * Required or optional status Use business-oriented names rather than raw URLs, element IDs, or API paths. ## Configure correlation Correlation determines which events belong to the same process instance. Potential identifiers include: * User session * RUM session identifier * Request or correlation identifier * Transaction identifier * Approved business-document identifier * Synthetic execution identifier * Application, environment, user, and timestamp ### Session-based correlation Use session correlation when the complete process occurs within one browser session. This is suitable for processes derived primarily from RUM events. ### Identifier-based correlation Use an approved transaction or document identifier when the process: * Continues across multiple sessions * Includes several users * Moves between browser and backend systems * Remains active for hours or days * Includes an approval workflow Do not collect sensitive identifiers unless necessary and approved. Mask or hash them when the original value is not required. ### Time-based correlation Time proximity can support analysis but is not reliable proof that two events belong to the same process. Use time-based correlation only when stronger identifiers are unavailable, and document its limitations. ## Configure alternative paths SAP processes frequently contain valid variations. Configure: * Optional steps * Role-dependent steps * Conditional approval * Rework loops * Rejection paths * Cancellation > **Show all 9** > > * Save and resume > * Multi-user handoffs > * Accelerated paths Do not classify every deviation from one ideal sequence as a failure. Example: ```text theme={null} Submit ├─ Approved ├─ Additional Approval Required ├─ Returned for Correction └─ Rejected ``` Each outcome can have a different completion or failure meaning. ## Configure abandonment A process is abandoned when it starts but does not reach an accepted completion condition within the configured period. Define abandonment according to the process. Examples: * Browser workflow: no further activity for 30 minutes * Checkout: no purchase completion within the attribution window * Approval: no approved, rejected, or cancelled state within the expected period * Long-running workflow: no state change within the configured SLA Do not mark an event sequence as abandoned merely because GermainUX lost monitoring coverage or could not correlate later events. Keep these outcomes separate: | Outcome | Meaning | | - | - | | Completed | Accepted completion condition occurred | | Failed | Configured application or business failure occurred | | Abandoned | Process started but did not complete in time | | Cancelled | User or system intentionally cancelled it | | Pending | Process remains within the allowed period | | Unknown | Available telemetry cannot determine the outcome | | Monitoring gap | One or more required steps were not observable | ## Configure performance objectives Define expected durations for: * Entire process * Individual steps * User activity * System wait * Approval wait * Integration processing A process can complete successfully and still be inefficient. Use representative baseline data before setting warning and critical thresholds. ## Analyze productivity GermainUX can analyze productivity using observed process and interaction data. Potential measures include: | Measure | Meaning | | - | - | | Process duration | Process duration | | Step duration | Step duration | | Completion rate | Completion rate | | Abandonment rate | Abandonment rate | | Failure rate | Failure rate | | Repeated steps | Repeated steps | | Rework | Rework | | User wait time | User wait time | | Excess duration above baseline | Excess duration above baseline | | Number of users affected | Number of users affected | | Frequency of inefficient paths | Frequency of inefficient paths | ### Lost productivity Lost productivity can be estimated by comparing actual duration with an approved baseline. For one process instance: ![Business Process Monitoring for SAP (Configure): Lost productivity](https://mintcdn.com/germain-ux/ZJflvRMC808oZDAs/images/main/business-process-monitoring-for-sap-configure/image-20260829-141851.png?fit=max\&auto=format\&n=ZJflvRMC808oZDAs\&q=85\&s=83455de0521f0e7c9a2f2217ff3fc472) Across multiple instances: ![Business Process Monitoring for SAP (Configure): Lost productivity, image 2](https://mintcdn.com/germain-ux/ZJflvRMC808oZDAs/images/main/business-process-monitoring-for-sap-configure/image-20260829-141903.png?fit=max\&auto=format\&n=ZJflvRMC808oZDAs\&q=85\&s=bde2e42a75b6b7d9f9aaae0228b44aa1) This is an estimate based on the configured baseline and observed events. It should not automatically be treated as employee performance measurement. ## Distinguish user time from system time Where the necessary browser events are available, separate: * User action time * User orientation or decision time * System or network wait time * Idle time The available breakdown depends on what RUM JS captures and how process steps are defined. Do not attribute the complete elapsed duration to application slowness or user inefficiency without supporting evidence. ## Configure errors within the process Associate relevant errors with the affected process and step. Use these classifications: | Classification | Meaning | | - | - | | User-facing application error | Software failure visible to the user | | Silent application error | Software failure not shown to the user | | User validation | Expected feedback preventing invalid input or action | | Business-process failure | Required business outcome was not achieved | | Integration failure | Configured downstream or upstream operation failed | Keep user validation separate from application errors. ## Analyze process segments Use pivots and filters to compare process performance by: * Application * Environment * Process * Step * User role * Department > **Show all 13** > > * Business unit > * Location > * Browser > * Device > * Release > * Completion outcome > * Error category Collect only approved user and business attributes. ## Configure outlier detection Outlier detection can identify unusual process or step durations after sufficient data has been collected. Potential use cases include: * A process takes much longer than its historical baseline. * One step becomes slower after a release. * A user segment follows an unusually repetitive path. * Failure volume increases unexpectedly. * A new process path appears. Outlier detection operates on collected process data. It does not expose steps that GermainUX cannot observe. ## Configure KPIs Recommended business-process KPIs include: | KPI | Description | | - | - | | Business Process | Complete process instance | | Business Process Step | Individual process step | | Process Volume | Number of process instances | | Completion Rate | Percentage reaching an accepted completion | | Failure Rate | Percentage reaching a configured failure | | Abandonment Rate | Percentage starting but not completing in time | | Process Duration | Total elapsed process time | | Step Duration | Elapsed time for one step | | Rework Rate | Percentage containing repeated or corrective steps | | Estimated Lost Productivity | Duration above the approved baseline | | Affected Users | Distinct users associated with the condition | Custom processes and steps can generate additional KPIs. ## Configure dashboards Go to: **Germain Workspace > Dashboards > All** Search for SAP or business-process dashboards, or create a dashboard using the configured process KPIs. Recommended views include: * Process volume * Completion and abandonment * Process-duration trend * Slowest steps * Step drop-off * Repeated steps > **Show all 10** > > * Error-affected processes > * Estimated lost productivity > * Segments with degraded performance > * Links to affected sessions and Session Replay ## Configure alerts and reports Create alerts for actionable conditions such as: * Process failure-rate increase * Completion-rate degradation * Abandonment-rate increase * Process or step exceeding its SLA * New error affecting a critical step * Unusual increase in rework * Synthetic process failure Schedule reports for: * Process efficiency * Completion and abandonment * Slowest steps * Process variation * Error impact * Lost-productivity estimates * Changes after a release ## Validate the configuration Test each supported path before production use. Confirm that: * The start event creates one process instance. * The completion event closes the correct instance. * Steps appear in the intended order. * Optional steps do not cause false failures. * Repeated steps are represented correctly. * Correlation joins only related events. > **Show all 13** > > * Failures use the correct classification. > * Abandonment timeout is appropriate. > * Monitoring gaps are not classified as abandonment. > * Process and step durations are accurate. > * Sensitive identifiers are protected. > * Synthetic traffic is distinguishable from real-user activity. > * Dashboards and alerts use the correct application and environment. ## Important limitations GermainUX business-process monitoring depends on the available telemetry. It does not automatically: * Discover every SAP workflow * Read undocumented SAP internal states * Correlate events without a shared identifier or validated rule * Determine business success from a browser click alone * Identify ABAP execution paths * Assign all elapsed time to either the user or the system * Measure native SAP GUI activity through RUM JS Additional SAP events may require custom RUM instrumentation, API access, log parsing, database queries, or another approved integration. ## Related documentation * [Configure Monitoring for SAP](https://docs.germainux.com/main/configure-monitoring-for-sap.md) * [Deploy Monitoring for SAP](https://docs.germainux.com/main/deploy-monitoring-for-sap.md) * [RUM JS Deployment for SAP](https://docs.germainux.com/main/rum-js-deployment-for-sap.md) * [Engine Deployment for SAP](https://docs.germainux.com/main/engine-deployment-for-sap.md) * [Configure User Monitoring and Session Replay for SAP](https://docs.germainux.com/main/user-monitoring-and-session-replay-for-sap-configu.md) * [Configure Error Monitoring for SAP](https://docs.germainux.com/main/error-monitoring-for-sap-configure.md) * [KPIs for SAP](https://docs.germainux.com/main/kpis-for-sap.md) ### Get More Information GermainUX can help determine which monitoring, analytics and automation capabilities are appropriate for your SAP environment. [Contact GermainUX Support](https://docs.germainux.com/main/support.md). ## Get Help > **[Contact GermainUX Support](https://docs.germainux.com/main/support.md)** > > The GermainUX Team can help you set this up. > **Can't find what you need?** > > Search all GermainUX docs with **Ctrl + K**, or use the feedback buttons at the bottom of this page to tell us what is missing. The content of this document is proprietary to GERMAIN SOFTWARE. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.