Amazon Quick Sight is a scalable, serverless, machine learning (ML)-powered business intelligence (BI) service. As a fully managed service, Quick Sight lets you create and publish interactive dashboards that can be scaled to hundreds of thousands of users without managing infrastructure.
As BI usage expands in your organization, you will have increasing demand of dashboards and have multiple authors contributing to dashboard creation. It’s important to identify who created a dashboard and when it was published. If there are issues in the existing dashboard, you can roll back to a previous version while fixing the current dashboard issues. Therefore, having features of version history tracking and the ability to re-publish dashboards is essential for effective platform management.
Amazon Web Services (AWS) recommends maintaining separate AWS accounts for different environments (Development, Testing, and Production) to ensure better security isolation. Implementing an automated deployment process using CI/CD pipelines helps maintain consistency and reduces human errors. See CI/CD and Quick Sight multi-account best practices to learn more.
However, if you haven’t yet established a multi-account environment and seek a no-code solution for dashboard versioning, you can learn an alternative approach in this post. You will learn how QuickSight’s dashboard versioning and publishing features can streamline your analytics workflow, from deployment to rollback. A sample sales performance dashboard will be used to demonstrate these features.
Quick Sight’s dashboard versioning and flexible publishing features offer the following benefits:
-
QA testing dashboards : You can create pre-production copies of analyses and dashboards for review, without touching your live reports.
-
Release history notes: When a dashboard is Published dashboard a version snapshot is created with version number, timestamp, author, and optional release notes. Each dashboard can store up to 1,000 versions.
-
Seamless dashboard updates: When publishing a dashboard, you can either create a new one or replace an existing one. Replacing an existing dashboard has several advantages over creating a new one. It preserves the same dashboard URL, user access rights, bookmarks and alerts. This means administrators don’t need to share a new dashboard URL or reconfigure user access, and users keep their bookmarks and alerts on re-published dashboards.
-
Accelerated dashboard rollbacks: You can quickly revert back to a previously published dashboard version with a few clicks in the console, making the rollback process faster.
Solution Overview
Let’s use this Sales Performance dashboard as an example. Notice the donut chart in the top right showing customer profit distribution. We’ll enhance this dashboard by converting the donut chart to a bar chart and implementing a dark theme.
Follow these steps to safely test changes in a staging environment, keeping your production dashboard intact. The following steps and screenshot outlines all necessary details:
1. Creating a Pre-Production analysis from Production Analysis
-
Open Production Analysis.
-
File → Save as Analysis to create a copy of the analysis.
-
Name it Pre-Production-Analysis and Save.
-
In Pre-Production-Analysis, update the analysis. In this example, we will change the donut chart to a bar chart and apply Midnight theme.
- After finished modification in Pre-Production Analysis, publish it to a Pre-Production Dashboard for QA testing by following steps.
2. Publish Pre-Production Analysis to Pre-Production Dashboard
-
Open Pre-Production Analysis.
-
Click PUBLISH on the right upper corner of the Pre-Production Analysis.
Select New dashboard, name it “Pre-Production-Dashboard.” Add release notes and PUBLISH DASHBOARD.
- Share the dashboard with Quality Assurance testers for testing.
After successful testing, deploy your changes by publishing the pre-production analysis to replace Production dashboard by using publish any analysis to any dashboard feature.
3. Publish Pre-Production Analysis to replace Production Dashboard
-
Open Pre-Production Analysis.
-
Click PUBLISH on the right upper corner of the Pre-Production Analysis.
-
Choose tab Replace existing dashboard → Browse all dashboards.
-
Choose Production Dashboard. Click SELECT on the right bottom corner
-
You can add release notes and PUBLISH DASHBOARD.
Once published, Quick Sight automatically creates a version snapshot of the dashboard. Dashboard owner and the co-owner can see the version information which includes version number, timestamp, author and release note.
4. View version history
To review the version history of your dashboard, follow these steps:
-
Open Production Dashboard.
-
Click Version history on the tool bar.
-
You can see the version information.
After each publish, Quick Sight snapshots the dashboard:
With the dashboard versioning feature, you can rollback to a previous dashboard version. This helps to minimize the impact of any issues that may occur with a newly released dashboard.
5. Instant Rollback
If you need to revert to a previous dashboard version, follow these steps:
-
Open Production Dashboard and open Version history.
-
From Version history, select the version you want to revert back.
-
Click PUBLISH and CONFIRM to restore it as the current dashboard.
-
Your dashboard will now be reverted back to previous version.
After reverting back your dashboard to a previous version, if you want to make changes on the dashboard using the rolledback version, you can save dashboard as a new analysis.
6. Save dashboard as a new analysis
To create a new analysis from an existing dashboard, follow these steps:
-
Open Production Dashboard.
-
Click Save as in the tool bar.
- Give a new of the new analysis. Then you can make changes in the new created analysis and publish it as dashboard.
Conclusion
In this post, we’ve demonstrated how Quick Sight’s versioning and publishing features revolutionize dashboard management.
You can now:
- Test changes with confidence in staging environments
- Track every dashboard iteration with detailed version history
- Update dashboards while preserving URLs and user permissions
- Restore previous versions instantly if needed
Start leveraging these powerful features today to streamline your BI operations and deliver a more reliable analytics experience. You can refer to these if you want to learn more about dashboard versioning and how to publish any analysis to any dashboard. You can also visit the Amazon Quick Suite Community to ask and answer questions, connect with peers, and stay updated on the newest features. Om the Community you can explore 500+ learning resources, weekly live online workshops, demos and more.
About the Authors
Roy Yung is a Specialist Solutions Architect for Amazon Quick Sight. Roy has over 10 years of experience implementing enterprise business intelligence solutions. Prior to working for AWS, Roy delivered BI and data platform solutions in the insurance, banking, aviation, and retail industries.
Snehali Mitra is a Cloud Support Engineer in Data Analytics at Amazon Web Services. She is a Subject Matter Expert in Quick Sight and Redshift platforms. She excels in helping customers with their Analytics Workloads and it is complemented by a genuine passion for business intelligence technologies.
Selma Rahmoune is an Associate Delivery Consultant at AWS specializing in Data analytics. With deep expertise spanning the entire data lifecycle, Selma architects and delivers end-to-end data platform solutions. Whether solid foundational data pipelines, sophisticated BI dashboard or anything in between, Selma delivers end-to-end data platform solutions for customers across several industries.