AWS Business Intelligence Blog

What’s new with data visualization in Amazon QuickSight – 2023 in review

Amazon QuickSight is a fully managed, cloud-native business intelligence (BI) service that you can use to connect to your data and create interactive dashboards that can be shared with tens of thousands of users. The dashboards can be used within QuickSight or embedded in software as a service (SaaS) apps.

As we bid farewell to 2023, let’s reflect on the exciting year of new data visualization launches in QuickSight key features that enhance data analysis for our users. This year was marked by significant advancements, including the infusion of generative BI capabilities for increased productivity among authors and the introduction of new visual features that facilitate the visualization of diverse data types in QuickSight. Our focus remained on elevating visual presentation and readability, achieved through a unified coloring experience and new KPI layouts. We optimized tables and pivot tables for space efficiency and interactivity, making it seamless to visualize and maintain context while navigating through larger tables. Furthermore, specific enhancements were tailored for KPIs, scatterplots, small multiples, and radar charts, supporting advanced use cases, along with axis improvements for enhanced readability. Notably, our strides in handling larger data volumes and improving performance at scale underscore our commitment to providing a comprehensive and efficient data analysis experience.

Author productivity improvement with generative BI

Creating compelling dashboards often involves hours of visual adjusting and refining by business analyst teams, altering visual properties using multiple point-and-click steps to achieve organizationally preferred presentation formats. The generative BI capabilities of QuickSight now allow business analysts to customize visuals to achieve a specific presentation using straightforward natural language prompts. Visual customizations can be specific in the build for each prompt, and several customizations can be included in a single prompt in order to quickly complete many visual editing tasks. The following screenshot shows an example dashboard. For more information, refer to Generative BI dashboard authoring capabilities now available in preview for Amazon QuickSight Q customers.

The following customizations are supported during this preview, with more coming in the general availability release:

  • Change visual type, such as “change to bar chart”
  • Change axis names and table column names, such as “rename Y axis to Account Manager”
  • Show or hide data zoom, such as “show data zoom”
  • Add fields to visuals or specific field wells, such as “add profit”
  • Change visual sort controls, such as “sort by sales descending’”
  • Apply conditional formatting, such as “make profits > 0 red”

For a full list of available customization options, refer to The generative BI authoring experience. This list will be updated as we add support for more operations.

New visuals

QuickSight dashboards now feature two new visuals: radar charts and geospatial heatmaps.

Radar charts

Radar charts (also known as spider charts, polar charts, web charts, or star plots) is a new visual type in QuickSight to help visualize multivariate data such as such as comparing cars across different stats like mileage, max speed, engine power, and driving pleasure. They are a great option when space is a constraint and you want to compare multiple groups in a compact space, and spot outliers and commonality. The following screenshot shows an example radar chart. Refer to Visualizing multivariate data using a radar chart in Amazon QuickSight to learn more about the various use cases and their data configuration.

Geospatial heatmaps

Additionally, authors can now improve readability of points on maps visuals by changing the point style to heatmap. This new way of displaying points on maps transforms individual points into a heat layer using color gradation, enhancing the readability of densely plotted, overlapping points. The geospatial heatmap style employs varying colors to signify areas of high and low data point concentration, facilitating identification of patterns, trends, and outliers. This new point style on maps not only improves readability but also maintains interactivity. Readers can still zoom in and out, pan across the map, and explore the data in detail. Additionally, when the map is zoomed in to a specific level, the heat layer seamlessly reverts back to the basic points. This transition enables readers to interact with the underlying points, including viewing tooltips and performing other actions based on the points. The following screenshot shows an example heatmap. Refer to Geospatial heatmaps in Amazon QuickSight for more details.

style=

Improved presentation

In this section, we discuss various presentation improvements in QuickSight.

Unified coloring with field-based coloring

With field level coloring, you can assign specific colors to individual dimension values across all visuals in a QuickSight analysis or dashboard. Colors are assigned on a per-field basis to simplify the process of setting colors and ensure consistency across all visuals that use the same field. Additionally, we are improving the coloring experience where visual colors will persist and not change with visual interactions like sorting, filtering, and actions. For more details, refer to Working with field level coloring in Amazon QuickSight.

New options for the KPI visual

The KPI visual has undergone a significant transformation with exciting updates:

  • Predesigned KPI layouts – The KPI visual now offers a user-friendly onboarding experience, allowing authors to select from predesigned KPI layouts tailored to various use cases and configurations. This empowers authors to effortlessly craft visually appealing KPIs with just a few clicks.
  • Sparklines – You can also gain insights into the trend of your KPIs over time by incorporating sparklines, which include line and area charts, in addition to progress bars.
  • Improved conditional formatting – Conditional formatting rules are now associated with specific fields, extending the capability to apply formatting rules independently to both the actual and comparison values, irrespective of which one is designated as the primary value.
  • Redesigned format pane – To accommodate these new additions, the format pane has been redesigned, enhancing navigation and simplifying the process of setting font colors for metrics directly from the format pane, eliminating the necessity for conditional formatting. For more details, refer to Using KPIs.

Data bars for tables

Data bars are bar charts displayed for a given column, where the bar length specifies the magnitude of the cell value relative to the range of values in the same column. Data bars make it straightforward to compare values and quickly spot outliers. With QuickSight, you can now use data bars on numeric fields and adjust your color scheme for both positive and negative values individually. For more details, refer to Adding data bars to tables in QuickSight.

Total and subtotal improvements for pivot tables

QuickSight now features improvements to totals and subtotals in pivot tables.

Row and column total positioning options

QuickSight now supports the total positioning option for both rows and columns in pivot tables, providing authors added flexibility to style and present tabular data according to your specific use cases and individual needs. Additionally, it makes it easier to compare and view totals that could otherwise be multiple scrolls or pages away. Authors have the flexibility to arrange row totals at the top or bottom, or set them to auto. The auto option is particularly useful for smooth transitions between compact and tabular layouts in pivot tables. In the compact layout, the default position is the top, whereas in the tabular layout, the default position is the bottom. Furthermore, authors can position column totals either to the left or right, with the default being the right position. To learn how to configure these options, refer to Totals and subtotals.

Contextual row subtotals

Instead of a generic subtotal label, authors can customize row subtotal labels by appending the group name to your subtotal in pivot table, enabling clearer association and understanding. For instance, if the current group is “North America,” you can rename your subtotals as “North America Total,” “Total for North America,” or “Total North America.” This feature is especially valuable for multi-page, multi-level reports, where subtotals lack meaningful insights if the associated group is not visible to the author or reader within the given page or view. Note that contextual subtotals are currently available in tabular pivot options. For hierarchy layout, the subtotals are part of the hierarchy levels itself. For additional details, refer to Totals and subtotals.

Custom totals

Authors now have the ability to define total and subtotal aggregations independently from the underlying aggregation. You can define the total and subtotal aggregations for your table or pivot table visuals directly from the field wells. For tables, the custom total menu is only available if totals are enabled. This feature allows you to select calculations like Sum, Average, Min, Max, or None (Hide) from a predefined list for computing totals in table and pivot table charts. This simplifies the process with a one-click method within the user interface, ensuring accurate data at both the row level and total calculation. For more information, refer to Custom total values.

Info icon for controls

The info icon empowers authors to provide clarity by adding extra details and context to dashboard controls, such as filters or parameter controls. Readers can easily access these supplementary notes by hovering over the controls, ensuring they better understand the control’s intended use and the data it represents. This feature effectively addresses the common challenge of control misunderstanding and misinterpretation, enabling authors to guide readers in comprehending specific controls or data nuances. For instance, you can specify the meaning of a control labeled “Time Span,” clarifying whether it refers to fiscal years, calendar years, or another time measurement.

Improved space optimizations

In this section, we discuss improvements to space optimization in QuickSight.

Hierarchy layout for pivot tables

The new hierarchy layout allows users to view data in a hierarchical manner, where all row fields are presented in a single column with indentation to differentiate between items from different fields. The hierarchy layout optimizes space by reducing the space required to display row fields, creating more room for numeric data and resulting in more organized and compact pivot tables. Additionally, the hierarchy layout supports enhanced navigation options through the updated pivot tables context menu. This provides users with a drill-down experience, enabling you to analyze root causes and explore data in greater detail for thorough analysis. For additional details, refer to Creating a pivot table.

Default column width

Authors can now effortlessly set a uniform column width for all value columns in your pivot table. You no longer need to perform the tedious task of manually adjusting individual column widths—this feature allows you to establish the desired width for all columns at once. The best part is that even after data refreshes, your column widths will remain consistent, with new columns automatically adhering to this default value. You can precisely set the column widths in pixels, giving you greater control over the presentation and ensuring a visually appealing layout.

Hide collapsed columns

The option to hide collapsed columns in tabular layout pivot tables empowers authors to manage the behavior of collapsed columns. This feature automatically hides collapsed row header fields, eliminating the necessity for excessive scrolling and contributing to a more streamlined and organized appearance of pivot tables. For further information, refer to Table and pivot table formatting options in QuickSight.

Visual-specific updates

QuickSight has also updated specific visuals, including scatterplots, small multiples, radar charts, and text boxes.

Scatterplot enhancements

The enhanced scatterplot visual now supports six additional use cases, including the ability to visualize unaggregated data across two dimensions: Color and the newly introduced Label field. With the new aggregate option called None on the field menu, you can plot unaggregated values even when using a dimension field on Color. In cases where one value is aggregated, the other value will be automatically set as aggregated, and the same applies to unaggregated scenarios. Furthermore, you can use the new Label field well alongside the existing Color field to add more flexibility in data visualization by allowing you to color by one field and label by another. Lastly, the performance of scatterplots has been improved to load up to six times faster, which applies to both new and existing use cases.

The following screenshot shows an example of unaggregated X and Y with Color and Label fields.

Small multiples axis improvements

The small multiples visual now offers new axis options for line and bar charts, granting authors the flexibility to tailor axes according to your specific use cases. You can now opt for shared or independent axis configurations for both the X and Y axes. The independent Y axis facilitates straightforward comparison of categorical value distributions across panels and supports the analysis of temporal data for period-to-period changes. This proves particularly beneficial when comparing data with varying magnitudes or scales, ensuring accurate and focused analysis of each individual chart without the influence of scaling in other charts. Similarly, for hierarchical dimensions like product or region hierarchy, an independent X axis allows for focused comparisons within a panel by displaying relevant X axis labels. This is especially useful for avoiding the display of blank or missing values with common axis settings. Furthermore, support for reference lines has been introduced, along with new label position options for both X and Y axes, providing additional customization capabilities.

Radar chart axis improvements

Similar to small multiples, the radar chart now supports shared and independent axis configuration. The shared axis option facilitates comparison across category axes, such as comparing test scores across different students. The Auto option dynamically determines the optimal scale, ensuring a smooth experience when transitioning between different chart configurations to explore data.

Pixel-based font size and text highlighting in text boxes

The text box visual has been enhanced to support pixel-based font sizing, providing authors with unparalleled precision. This improvement enables you to adjust font sizes based on numeric values (such as 14 or 32), offering greater flexibility in text presentation. Additionally, we have added text highlighting capabilities, similar to that of the Insights visual. You can highlight text to emphasis key data points and insights, making it easier to drive user attention precisely where you want it, and making your data-driven narratives even more compelling. For more details, refer to Using text boxes.

Interactivity by freezing columns

Both authors and readers can freeze columns to affix columns in place while scrolling horizontally, ensuring it remains visible and accessible even as other columns move out of view. The frozen column is moved left until it reaches another frozen column, maintaining its fixed placement. You have the flexibility to freeze and unfreeze a single column, multiple columns individually, or even a group of columns together using the Freeze up to option. Freezing columns enhances visibility, context, and access to critical information, making it easy to navigate through extensive tables without losing sight of important details.

Increased data limits and performance

QuickSight has also increased its data limits and improved performance in various ways.

Performance upgrades and increased data handling for pivot tables

Pivot tables now boast increased data handling capabilities, supporting more extensive datasets to unlock advanced use cases. Notably, the field limits have been raised, with the value field well limits growing from 20 to 40, and both rows and columns expanding from 20 each to a combined limit of 40. For instance, if you have 34 fields in the rows, you can add a maximum of 40 – 34 = 6 fields to the column field well.

To accommodate these expanded field limits, performance enhancements have been implemented to load pivot tables faster. Instead of fetching the entire dataset in the viewport, the system now fetches data only for visible fields (expanded fields) along with a small subset of values under the collapsed field. This makes sure that data fetched in every new query will be used to render new values that can be displayed immediately.

As a result of these optimizations, customers have reported improvements in load times ranging from 2–10 times faster, depending on the complexity of their dataset.

New scatterplot data limits (soon to be generally available)

We are soon to launch flexible data limits for scatterplots, allowing you to view your complete data or maximum of 2,500 data points, whichever is less in case of both unaggregated and aggregated scenarios. This will allow you to access and override our default field limits, enabling you to view your complete data.

Performance improvements for histograms

Significant improvements have been made to the performance of histograms, resulting in loading times that are 3–4 times faster. This enhancement is particularly beneficial in preventing query timeouts when dealing with very large datasets exceeding 100 million records.

Conclusion

In 2023, QuickSight made significant strides in data visualization. The integration of generative BI streamlined visual creation and customization, and new visual types like radar charts and geospatial heatmaps expanded data representation possibilities. Improved presentation features, such as unified coloring, revamped KPI visuals, and space optimization, prioritize user-friendly layouts. Specific enhancements, including data bars and improved pivot table positioning, provide flexibility and control. Visual-specific updates, like scatterplots and small multiples improvements, showcase dynamic capabilities. Interactivity has been improved with column freezing in tables, and performance upgrades cater to larger datasets. 2023 encapsulates a commitment to innovation, user experience, and performance, ensuring QuickSight remains at the forefront of intuitive and powerful data visualization.

Feel free to share your questions or feedback by leaving a comment. For further discussions and assistance, explore the QuickSight Community. Stay informed about our latest launches by referring to the What’s New feed for QuickSight. Don’t forget to subscribe to the QuickSight YouTube channel for the latest workshops, how-to walkthroughs, and demo videos to stay up to date on all things QuickSight.


About the Author

Bhupinder Chadha is a senior product manager for Amazon QuickSight focused on visualization and front end experiences. He is passionate about BI, data visualization and low-code/no-code experiences. Prior to QuickSight he was the lead product manager for Inforiver, responsible for building a enterprise BI product from ground up. Bhupinder started his career in presales, followed by a small gig in consulting and then PM for xViz, an add on visualization product.