Looker Studio Charts Guide
Looker Studio (formerly known as Data Studio) charts are explained with an example and use case scenario below.
Check the live dashboard charts guide here →
Scorecard: Scorecards display a summary of a single metric and are the most commonly used way to visualize core key performance indicators (KPIs).
Scorecards
If your numbers are within 5 digits and are sensitive data like price and sales, where you want to see the absolute number
Recommended (❶): Yes, especially if you make reports for senior management and usually want to see up and down for the most important KPIs.
Scorecard with compact number
Recommended (❶): Yes, If your numbers are more than 5 digits and are not sensitive like Sessions and Page Views.
Scorecard for % Data
Recommended (B): Yes, only with absolute changes in configuration
How to create a unified scorecard?
Stacked on top of each other to display unified Absolute Changes and % Changes in a scorecard. Create a scorecard in 2,1 orders (2nd scorecard first) to avoid manually arranging stacking orders.
Recommended: Yes, with this format, you can create a cleaner design and also fit many scorecards in a row.
Padding and Line Height
Padding: Reserved space for Left, Right and Top
Recommended: Yes, but only in the particular case when you have to fit many scorecards in a tight space
Line Height: Space between rows in pixels.
Recommended: No, please leave it to default auto; this might condense two rows of data in one. PDF export will also result in a messy, disorganized result.
A Pie/Doughnut chart expresses a part-to-whole relationship in a data set. It will always add up to 100%. So, if, for instance, you want to show "How does 100% divide up into multiple shares?" you should always use a Pie/Doughnut Chart with % as a data label.
Pie Chart
If you want to show "how 100% divide up into a few shares?" You should always use Pie/Doughnut Chart with % as a data label.
Recommended (A): Yes, one of the most widely used charts across the industry
Doughnut Chart
If you want to show "how 100% divide up into a few shares?" You should always use Pie/Doughnut Chart with % as a data label.
Recommended (A): Yes, my personal favorite is to use the Doughnut chart; the advantage of using a doughnut chart has a space in the middle for a Scorecard. So you can know "What's the total number that makes 100%?"
Pie/Doughnut Chart
In this example, we can only guess the % share of any device.
Recommended: No, If you want to see a part-to-whole relationship in a data set. Charts data labels should be always % and avoid having number label.
Column/Bar Chart as a Pie Chart Alternative?
Bar/Column charts have different use and shouldn't be used as a Pie chart alternative.
Recommended: No, If you want to show % share within a KPI, please avoid the column or bar charts.
Bullet charts display a single metric performance and progress towards a certain goal or against a benchmark. Without a target, it can be also configured to display the ranges for poor, average and good. Check the example below.
Bullet Chart: With Target
Ranges are set manually, and the following range value should be greater than the last one.
Target is also set manually which is an optional metric. In this chart, the target revenue is set as 4 million, and the blue line represents how far the target has reached so far
As of Nov'22, It is not possible to show the data label in a bullet chart, but you can use a scorecard to display the absolute number.
Recommended: Yes
Bullet Chart: Without Target
Ranges are set manually, and the following range value should be greater than the last one.
In this example, we have not set a target value, and It can use to consider revenue up to 2 million as a poor result, above 2m and up to 3m as the average result, and above 3m as a good result.
Recommended: Yes
Bullet Chart: With Target and Compared to the previous period
Ranges are set manually, and the following range value should be greater than the last one.
In this chart, the target revenue is set as 4 million, and the blue line represents how far the target has reached so far. Light blue displays the revenue of the last year during the same period.
Recommended: Yes
Gauge charts display a single metric performance and progress towards a certain target/goal or against a benchmark and optionally divide them into up to 5 ranges.
Default Gauge Chart
Recommended: No, by default, a Gauge Chart doesn't provide any meaningful information without a Target or Ranges.
Gauge Chart With Target Only
This is used to measure a metric performance and progress towards a certain target or against a benchmark.
Recommended: Yes
Gauge Chart With Ranges Only
This can be configured to display the ranges for poor, average, and good. Up to 5 ranges are allowed as it shown in the next example
Recommended: Yes
Gauge Chart With Target and Ranges
This is used to measure progress towards a certain target/benchmark and divide them into 5 ranges.
Recommended: Yes
Gauge Chart With Target, Ranges, and Comparison
To measure a metric performance against a target and also compare it to the previous period and divide them into 5 ranges.
Recommended: Yes
Column Charts shows comparisons among discrete categories. One axis of the chart shows the specific categories being compared, and the other axis represents a measured value. The bars can be plotted vertically or horizontally.
Stacked Columns represent multiple types of data within a single bar stacked vertically or horizontally, one on top of another. The value determines the length of each series in each data point.
Column Charts: Single Axis
The number of pubs/bars should not be too large (the labels of the axis may appear incomplete).
Recommended: Yes, and keep the number of pubs/bars to a max of 15.
Leave to the default if X-Axis is the date or small data points "Day of Week" in this chart)
Column Charts with Multiple Metrics: Single Axis
Recommended: Yes, and keep the single-axis configuration in order to show the right comparison between the data
Column Charts: Dual Axis (1/2)
Suitable If the difference is not large or displaying % of total data (next example)
In this example, even though Conversion is lower than Total users, bars appear bigger. An example below is if we change the "Absolute Value" to "% of Total". Both Users and Conversions display correctly.
Recommended: No, charts appear wrong
Column Charts: Dual Axis (2/2)
Suitable If the difference is not large or displaying % of total data (next example)
In this example, we have changed "Absolute Value" to "% of Total." Thus, both Users and Conversions display correctly.
Recommended: Yes
Stacked Column: Single Axis
Like the above example, we will have the same wrong visual if we use a double-axis configuration. Keep the single-axis configuration.
Recommended: Yes, but only works for a few occasions where you want to compare the part of the total data. In this chart, "what were the New Users against Returning Users?"
100% Stacked Column: Single Axis (1/2)
Even though 100% stacked columns always equal 100%, it's best to have % data within the chart so you can see the actual % by showing the data label, else it's hard to visually compare (Unless the absolute value is necessary in your case) the relative size of the components.
In the next chart, we will display the same data as "% of Total".
Recommended: Yes
100% Stacked Column: Single Axis (2/2)
In this chart, we are displaying the above data as "% of Total", and become human-friendly to see the top performing device share each day.
Recommended: Yes
Combo charts use both axes to show a comparison between different KPIs. They work in a similar way to Column charts except with the option to configure "How to display individual metric?" (Line and Pub/Bar). The right axis is often used, If a data set is significantly smaller than the rest of the KPIs.
Combo Chart: Single Axis
A Single Axis configuration works well if the metrics you are comparing have at least a 3-5% difference. For example, if we compare Users and Revenue per User (small data) in the same chart, the Revenue per User data will be a straight line at the bottom of the X-Axis.
Recommended: Yes, Unless X-Axis is a date, it’s recommended to keep the number of pubs/bars to a maximum of 15.
Combo Chart: Dual Axis
The right axis is often used If a data set is significantly smaller than the rest of the KPIs, for example, Revenue per User shown in this chart
Recommended: Yes, always use double Axis configuration, If one of your data sets is significantly smaller than the rest of the KPIs.
Stacked Combo Chart (1/2)
In this example, we are comparing Revenue per User (currency) with users (raw number). In the stacked combo chart, a axes should have similar kinds of data, in order to have the right data type comparison
Recommended: No, this chart doesn't have the right data type comparison
Stacked Combo Chart (2/2)
A Double Axis configuration works well If one of the metrics you are comparing has a different kind of data set.
Recommended: Yes, we have moved Revenue per User to the right axis, in order to separate different kinds of data on the different axis.
Tables represent quantitative information where precise values can be required in some cases. However, this is not exclusive in this information age. My personal favorite method of using tables - use them as raw data export charts into Microsoft Excel, Google Sheets or CSV for detailed analysis or custom charts.
Table Chart
Tables can handle large data sets, and in many cases, Tables are the most suitable for displaying a large data set. Please use the chart type that best suits your need.
If all the rows are not visible in the table, display the row number and page navigation, so the audience can navigate easily if the table has hidden rows
Recommended (A): Yes, see the chart type that suits your need
Table With Bars
The "Table with bars" is very similar to the "Table chart" except that the "Table with bars" has the option to display the value as horizontal bars. It’s possible to display just Bars or both Bars with Numbers.
Recommended (A): Yes, see the chart type that suits your need
Table With Heatmap
A "Table with heatmap" is also similar to a "Table chart" except for the heatmap, which adds additional visual comparison where the highest value will have the highest color contrast.
Recommended (A): Yes, see the chart type that suits your need
Personal favorite: use them as raw data export charts into Microsoft Excel, Google Sheets or CSV for detailed analysis or custom charts.
[Optional Dimension & Metrics] — these are configurable in the DATA tab. Select the "Drill down" checkbox for Dimensions and "Optional metrics" for Metrics. You can extract/visualize the data with a combination of metrics and dimensions, including the optional.
[Optional Dimension: Drill up and down] — in a table, you can also add optional dimensions, for example, Date, Month, City, and Channels. Once you drill up or down (depending on the setup), you can change from County to any optional dimension you have set.
[Optional Metrics] — these work in the same way as optional dimensions. If it's not possible to fit all the metrics in your Table, use Optional metrics.
Recommended: Yes, a personal favorite — for raw data for export into Excel/Google Sheets for a custom chart or details analysis.
A Pivot Table stores the summary of a specific data set in a condensed manner or groups data together in a meaningful way. My personal favourite method is to treat Pivot Tables as raw data export into Excel and Google Sheets for detailed analysis.
Pivot Table
Pivot Tables can handle large data sets, and, in most cases, Pivot charts are the most suitable choice for displaying and summarizing an extensive data set.
Select "Show grand total" for both Rows and Columns in the data tab to show the total number for individual rows and columns.
Displaying % Changes is not allowed in Pivot Table inside Google Data Studio
Maximum 5 Pivot Tables are allowed on every page in Google Data Studio
Recommended (A): Yes
Pivot Table With Bars
The "Pivot Table with bars" is very similar to the "Pivot Table" except that the "Pivot Table with bars" has the option to display the value as horizontal bars. It’s possible to display just Bars or both Bars with Numbers.
Recommended (A): Yes
Pivot Table With Heatmap
Pivot Table with heatmap: A "Pivot Table with heatmap" is also similar to the "Pivot Table chart" except for the heatmap, which adds additional visual comparison where the highest value will have the highest color contrast.
Recommended (A): Yes
An Area Chart is best used to show the distribution of categories as parts of a whole changing over an interval. Plotted inside area chart is used to convey whole numbers, so they do not work for negative values. The number of series in an area chart should be kept moderate as too many series may well hinder its legibility.
Area Chart
Y-Axis plotted individually on each interval. In this chart, the smallest series area is plotted at the bottom. When hovering over the chart will display raw number for individual series at each interval.
The number of series in an area chart should be kept moderate as too many series may well hinder its legibility.
Recommended: Yes, see the chart version that suits your need
Area Chart: Show Stack
Y-Axis is plotted against the sum on each interval. In this chart, the smallest series area is plotted on the top. When hovering over the chart will display the Total Value together with raw data for individual series at each interval
Recommended: Yes, same as Area Chart, except in this chart, the smallest series area is plotted on the top rather than the bottom.
Area Chart: 100% Stacking
A stacked Area Chart must have a Breakdown Dimension, in this example, the Device Category
Recommended: Yes, if you want to split the trend over time into % share at each interval.
A scatter chart will show the relationship between two different variables represented by the X and Y axes.
The Bubble Chart is a perfect alternative for Scatter Chart and I always use a Bubble Chart instead, which has an option to display the third value in form of bubbles size. See the next example
Scatter Chart: With Bubble Color
With Bubble Color configuration each data point will have unique color by default which can be also configured to personal preference.
Recommended: Yes, test for the number bubble that can fit inside the chart, and too many bubbles will make the chart difficult to read.
It’s best to stay one-dimensional with both Bubble and Scatter Charts. More than one dimension will be hard to differentiate for most of your report’s audience.
Scatter Chart: Without Bubble Color
Use Bubble Charts: for the same configuration, bubble size will represent the third value and are the best alternative for Scatter Charts.
Recommended: No, all the bubble color appear the same and it's hard to differentiate data series unless you hover over the bubble.
Scatter Chart: Two Dimensions
We have two dimensions in this example, Channels and Device Category. If you have more than one dimension, the first selected dimension of the chart will be displayed as a Legend. As it is shown in this picture
Recommended: No, unless it's absolutely necessary, avoid having Scatter and Bubble charts with two dimensions.
If we apply the data label in this chart becomes messy and also it's hard to differentiate between two data points as the bubble's color appears the same. In this example, I have highlighted the Organic Search users for both Mobile and Desktop which has the same bubble color.
Bubble Chart = Scatter Chart + Bubble Size Configuration
A Bubble chart shows the relationship between two different variables represented by the X and Y axes, with Bubble Size (optional and without it will be a Scatter Chart) representing the relative size of the metrics values
Bubble Chart: With Bubble Color
With Bubble Color configuration each data point will have unique color by default which can be also configured to personal preference.
Recommended: Yes, always show both axis and test for the number of bubbles that can fit inside the chart, too many bubbles will make the chart difficult to read.
It’s best to stay one-dimensional with both Bubble and Scatter Charts. More than one dimension will be hard to differentiate for most of your report’s audience.
Bubble Chart: Without Bubble Color
Without color configuration, all the bubble's color will be the same and if you have many data points applying a data label make charts super crowded.
Recommended: No, all the bubble color appear the same and it's hard to differentiate data series unless you hover over the bubble.
Bubble Chart: Two Dimensions
If you have more than one dimension, the first selected dimension of the chart will be displayed as a Legend. As it is shown in this picture
Recommended: No, unless it's absolutely necessary, avoid having Bubble and Scatter charts with two dimensions.
If we apply the data label in this chart becomes messy and also it's hard to differentiate between two data points as the bubble's color appears the same. In this example, I have highlighted the Organic Search users for both Mobile and Desktop which has the same bubble color.
Maps: Geo Charts are an easy way to visualize data in maps to compare how a measurement varies across a geographic area. The dimensions are related to continents, countries, regions/states and cities, with your desired metrics as a data point.
Geo Chart
Chart areas are color-filled based on High, Medium, Low and data-less values. A Geo Chart only has the option to set Zoom Area.
Geo Chart only has an option for Zoom Area, in the chart B, Zoom Area is set to the United States.
Recommended: Yes, [a personal favorite]: see the chart version that suits your need
Filled Map
Almost the same as the Geo chart, except that there is no option to set the Zoom area. Filled maps have two background layer options (map and satellite views) which can be configured in the Style Tab.
Recommended: Yes, see the chart version that suits your need.
Bubble Map
Chart areas are filled as Bubbles and the chart has an option to display as a Map or Satellite view. Which can be configured in the Style Tab. You can also set the number of Bubbles and the Bubbles Size.
Recommended: Yes, see the chart version that suits your need.
Heat Map
Same as other geo charts except you have the additional options which include Heatmap aggregation as SUM or MEAN; Min and Max color domain value, and Intensity of the heat.
Recommended: Yes, see the chart version that suits your need.
Treemaps: Treemaps show parts of a whole and display hierarchical information.
Treemap Chart: Single Dimension
Colour depends on the value of the metric and can be configured within the style tab. In this chart, the max colour value is set to Green, Mid = Light Yellow, and Min = Light Red.
Max Row: Data Points to show within the Chart, excluding the headers
Recommended: Yes, With A Single Dimension Configuration
Treemap Chart: Two Dimensions
If you have to have this chart test the number of rows that can fit nicely inside your graph and display the chart on the full-width page, for example below
Max Row: Data Points to show within the Chart, excluding the headers
Recommended: No, In this example, over 50% of revenue is coming from the US and the chart looks great, but a treemap map can be hard to visualize for two dimensions. Avoid having two dimensions, unless it's absolutely necessary.