Build pixel-perfect reports with ease using Amazon Q in QuickSight

Amazon recently announced that the generative AI capabilities of Amazon Q are now generally available in Amazon Quicksight. At the core of Amazon Q in QuickSight are several powerful capabilities that harness the power of natural language processing:

Create visuals in seconds using the new Amazon Q powered authoring
Fine-tune and format visuals using natural language prompts
Generate calculations without needing to know specific syntax
Craft compelling visual narratives and customize them with ease

Building on this innovation, QuickSight users can now use natural language generation to create pixel-perfect reports – a new capability in Amazon Q for QuickSight that simplifies the way users create and distribute visually-rich, highly-formatted reports to their stakeholders. Pixel-perfect reports are crucial because they ensure that every detail of the report, from layout to formatting, is meticulously controlled and aesthetically pleasing. This not only creates a professional data presentation, it also improves comprehension and decision-making among stakeholders.

With the extension of these generative AI capabilities to pixel-perfect reports in Amazon Q for QuickSight, the time-consuming and manual effort traditionally required to create reports can be significantly reduced, allowing organizations to focus on the substance of their reports and communicate critical business data more effectively.

Moreover, these capabilities seamlessly integrate with the robust security features found in QuickSight, ensuring that row-level and column-level security rules are applied to the data sets used in the reports. This means that the same report can be delivered to multiple recipients without compromising sensitive information or the need to create separate reports for each individual.

In this post, we cover how to use generative BI capabilities to accelerate pixel-perfect report designing using Amazon Q in the QuickSight console and deliver it to QuickSight users. Pixel-perfect reports can also be delivered to non-QuickSight users in Excel, CSV, or PDF format using Snapshot Export APIs.

Use Case: Assist business users accelerate the design and development of pixel-perfect loan analysis reports using Amazon Q in QuickSight

In this use case, we will explore how a loan officer can generate pixel-perfect reports for a financial organization to analyze load data across their customers. This assists loan offers and relevant stakeholders by enabling timely decision-making for loan applications, approvals, repayments, delinquency tracking, portfolio summaries, credit risk analysis, and more. This approach can also be applied to various domains, industries, and personas.

Solution overview

The following diagram illustrates high level architecture of pixel-perfect reports using Amazon Q in QuickSight.

Prerequisites

To follow along with this example, you will need to have a QuickSight Enterprise account. In order to have pixel-perfect reporting capabilities, your organization needs to subscribe to the pixel-perfect report add-on within the account. For subscription pricing, refer to Amazon QuickSight pricing.

Amazon Q is automatically enabled for accounts with at least one Pro user or with at least one Amazon Q topic.
Generative BI capabilities with Amazon Q in QuickSight may not be available in all AWS regions as of this writing. For this post, we use the US East (N.Virginia) Region.

Upload your dataset

For this post, we use a sample dataset, customerloans.csv. Complete the following steps to upload your dataset:

On the QuickSight console, choose Datasets in the navigation pane.
Choose New dataset.
Choose Upload a file.
Browse to the customerloans.csv file and select Open.
In the Confirm file upload settings pop-up window, choose Next.
Choose Edit/Preview data.
In the Data Preparation pane, choose Save & Publish.

Your dataset is now imported and ready for use.

Create a topic

Complete the following steps to create a QuickSight topic:

On the QuickSight console, choose Topics in the navigation pane.
Choose New topic.
Name the topic Customer Loans, include a relevant description, and choose Continue.
Choose the dataset you want to use to create the topic, and choose Continue button.

Amazon Q indexes the data and sets up field configurations as part of the topic creation. This can take a few minutes. The status column shows the progress of this task.

Create a pixel-perfect report sheet

Complete the following steps to create an analysis:

On the QuickSight console, choose Analysis in the navigation pane.
Choose New Analysis.
Choose the CustomerLoans dataset as the dataset to include in your analysis.
Choose Use in analysis.
Choose the pixel-perfect report option.
Keep the default settings for paper size and orientation.
Choose Create.

You are now in an analysis view. This is where you can add visuals, adjust the layout, and publish your report. First, let’s change the name of the analysis.

Choose the name of the analysis to rename it.

Link your topic to the analysis

Complete the following steps to link your topic:

Choose the options menu (three vertical dots) next to Build visual and choose Topic Linking.
Enable Link topic for Build Visual and Q&A.
Select the Customer Loans topic.
Choose Apply Changes.
Exit the Topic linking modal.

Add a calculated field with Amazon Q

For this example, we want to calculate the quarter of the year based on an input date. To create a calculated field using natural language, complete the following steps:

Choose Calculated field Data pane.
Enter Quarter for the name of the calculated field.
Choose Build Calculation to use natural language prompting.
Enter quarter given issued on date into the text field and choose Build.

Amazon Q will construct a calculation and present you with the results for approval.

Choose Insert to approve and use the generated calculation.
Choose Save to save the calculated field.

The calculated field is now available for use in the analysis.

Build a new visual with Amazon Q

Now we want to build a new visual. Complete the following steps:

Choose Build visual.
Enter build table with cust id, first name, last name, age, annual income, loan term and choose Build.

Amazon Q will generate a visual to represent your query.

Choose Add to analysis and close the Build a visual modal
You can position the newly created visual as desired.

Edit a visual with Amazon Q

Lastly, we want to change the title of the visual. To edit your visual, complete the following steps:

Choose Edit visual.
Enter change title to Loan Applications and choose Apply.
To update the column titles, enter change Cust Id to Customer ID and choose Apply.
To add the Purpose field as a column in the table, enter add Purpose and choose Apply.

You can perform additional operations such as changing the visual type, as well as conditionally formatting your data. For a complete list of Amazon Q capabilities for editing visuals with natural language commands, refer to Refine visuals with generative BI.

Publish and schedule your report

Now you are ready to publish your report. Complete the following steps:

Choose Share and Publish dashboard.
Name the dashboard “Customer Loans Report”.
Choose Publish dashboard.
You can then either generate the report as a PDF and download it, or schedule to send the report to recipients in PDF, CSV or Excel format at your desired date and time.

The following screenshots show examples of our sample report.

Sample Report

Clean up

If you want to stop using this add-on and Amazon Q capabilities in Quick Sight and avoid monthly recurring fees, complete the following steps:

Unsubscribe from the pixel-perfect reporting add-on.
To deactivate Amazon Q, remove all Pro users and topics.
Delete the data set to release SPICE capacity.

You will be charged for the current month’s subscription. For subscription pricing, refer to Amazon QuickSight pricing.

Conclusion

In this blog, we went over a use case of a financial organization looking to analyze loan data across their customers and enable timely decision-making. We used generative BI authoring capabilities powered by Amazon Bedrock to design a pixel-perfect report, thereby accelerating report design and reducing development efforts from days to hours or even minutes. Generative BI can help minimize the need to have deep coding experience or knowledge of syntax to design reports. Although this post was focused on customer loan analysis and the financial domain, you can apply this solution to a number of scenarios and various types of documents or reports.

Take control of your business-critical operational reporting with Amazon QuickSight pixel-perfect reports. This fully managed cloud solution streamlines report creation, scheduling, and sharing across your entire organization. Design pixel-perfect, multi-page reports that share your analytics in a professional, print-ready format. Customize each page with layouts, images, and data visualizations from QuickSight. Then distribute your reports in PDF, CSV or Excel format for viewing by colleagues or stakeholders. Start using the power of pixel-perfect reports in Amazon QuickSight today! To learn more, refer to Amazon QuickSight pixel-perfect reports, get hands-on with a workshop, and dive deeper into the details.

References

About the author(s)

Deepak Sahi is a Solutions Architect at AWS based out of Zurich, Switzerland. He has close to two decades of experience in the field of data analytics, primarily in business intelligence and data warehousing. He has worked worldwide as a consultant in domains such as telecom, finance, insurance and healthcare for many Fortune 500 companies. He is currently focusing on manufacturing companies in Switzerland and helps them build secure and innovative cloud solutions, enable data-driven decisions and solve their business challenges.

Dinesh Thakkar is a Senior Technical Account Manager and Analytics Technical Field Community Member at AWS with over 21 years of experience in IT. He currently supports a global financial services customer and provides consultative architectural and operational guidance to Enterprise Support customers to help them achieve the greatest value from AWS.

Aashi Agrawal is a Solutions Architect at AWS, where she specializes in the analytics domain. She guides customers through the transformative process of migration and modernization. With a blend of visionary architecture and robust security, she crafts resilient systems and seamlessly integrates cutting-edge AI/ML services, including the marvels of generative AI, into their technological tapestry. Outside of work, she loves to explore new things and discovers music.

Aneri Modi is an AWS Solutions Architect based out of Pennsylvania. She currently works with higher education customers to architect scalable and robust cloud solutions. She specializes in Analytics and AI/ML technologies.

Kiki Nwangwu is an AWS Associate Specialist Solutions Architect. She specializes in helping customers architect, build, and modernize scalable data analytics solutions.

Michael Purpura is a Senior Enterprise Solutions Architect based out of Irvine, California. He enjoys working with customers to realize the transformative value of data-driven operations, with a passion for bringing data to life through business intelligence.

AWS Business Intelligence Blog