- Release Notes Cloud Insights
- Getting Started
- About Insights
- Prerequisites
- Building automations for Insights
- Insights Data Model
- Access and Permissions
- Notifications
- Interacting with Insights
- Action Center Integration
- Automation Hub Integration
- Autopilot integration
- Document Understanding Integration
- Healing Agents integration
- License monitoring integrations
- Maestro Integration
- Test Manager Integration
- Real-time Monitoring
- Real Time Data Export
- Licensing
- Troubleshooting
Insights
When designing automations, it is important to consider not only execution but also how those automations will be measured and reported in UiPath Insights. Reporting requirements are often overlooked during automation design, which can make it difficult or sometimes impossible to track key metrics later without reworking the automation.
This page outlines best practices to ensure your automations provide the right data to Insights from the start.
- Missing data points that prevent accurate reporting (for example, the number of transactions processed per job).
- Non-standardized log messages that make reporting unreliable and inefficient.
- ROI reporting gaps, where Insights cannot determine the volume or value of work processed.
For example, if an automation processes an Excel file where each row represents a transaction, but the job does not use queues or log these transactions via custom variables, Insights cannot accurately calculate cost savings or ROI.
1. Define reporting requirements at design time
Before building, identify the KPIs and reports you and your stakeholders will need:
- ROI metrics (for example, number of transactions processed, time saved).
- Business outcomes (for example, orders completed, invoices processed).
- Operational health (for example, queue backlog, success vs. failure rates).
Make sure your automation is designed to capture these data points in a structured way.
2. Use queues for transaction-based automations
If a job processes multiple transactions:
- Use Orchestrator queues to represent each transaction individually.
- If queues are not feasible, avoid grouping transactions together without logging the number of transactions processed.
This ensures that Insights can measure throughput, success/failure rates, and processing times per transaction.
Leverage custom variables
Custom variables allow you to bring business-relevant data in Insights for reporting. Examples include:
- Invoice amount
- Customer ID
- Priority level
- Processing cost
By mapping these variables during automation design, you can unlock richer reporting in Insights.
When designing automations intended for Insights reporting:
- Define reporting requirements up front.
- Use queues for multi-transaction jobs.
- Add custom variables to capture key business data.
By following these practices, you make sure that Insights can provide accurate, meaningful reports without requiring rework later.
