You are here:
Deployment Insights
Compare org performance before and after deployment to identify regressions introduced by recent changes.
Considerations
- Deployment data is available starting 4 days after your org opts in to Scale Center.
- Deployments from the last 30 days (up to yesterday) are available for analysis.
- If no deployments appear, check back after you run a deployment.
- Wait at least 24 hours after deployment completion before generating a report.
- Deployment Insights analyzes a 7-day window: the 3 days before the deployment, the deployment day, and the 3 days after. If fewer than 3 days of post-deployment data are available, the report uses the maximum available time window.
- Peak hours are identified based on request volume spikes. If your org has low weekend traffic, peak hours are selected from high-activity weekdays instead to ensure accurate before-and-after comparisons. If your deployment occurs before a weekend, the analysis uses peak hours from high-traffic weekdays rather than low-activity weekend periods.
- All insights use the same peak-hour methodology, including errors, user experience, API performance, and infrastructure regressions.
To generate a deployment insights report:
- From Setup, in the Quick Find box, enter Scale Insights, then click Deployment Insights.
- Select a deployment from the last 30 days (up to yesterday).
- Click Generate Insight Report.
| Section | Description |
|---|---|
| Successful Deployments and Failed Deployments | Displays a summary of deployment results over the last 30 days. |
| Change in Errors Before and After Deployment | Compares peak hour error counts from before and after deployment to help you identify whether the deployment increased or resolved errors. |
| Change in Performance Before and After Deployment | Compares performance metrics from peak traffic hours before and after deployment to assess the impact on your org. |
| Additional Insights on Errors, User Experience, API Performance, and Infrastructure Performance Regressions | Provides detailed information on regressions in these areas, including telemetry on how to optimize performance for recent deployments. |

