Trends show the evolution of your variables over a date range. There are currently 4 different types of trends chart, each visualising your data in a different way.
The Trends panel of your project is a space to inspect patterns across time in your dataset. You can quickly plug in variables and segmentations from your analysis to plot, compare and correlate them with other variables.
In trends charts, date variables are plotted on the x-axis and quantitative variables are plotted on the y-axis. You can change which variables are shown using the variable dropdowns at the top of the Trends panel.
"There is magic in graphs. The profile of a curve reveals in a flash a whole situation - the life history of an epidemic, a panic, or an era of prosperity."
- Henry D. Hubbard
You can create 4 types of charts in your project's Trends panel. Each represents your data differently. It's easy to switch between types of charts after you've created a chart.
Overview charts are histograms representing all of your data. Here you can see how variables from your data evolved over time.
Best used for ... Investigating general trends.
Segmented overview charts show how values from specific variables evolved over time. These are represented as a series of histograms, each representing a different value belonging to a specific variable.
Best used for ... Inspecting how trends differ between values of a variable.
Compared segment charts represent your data in a line graph. Similar to segmented overviews, they plot different values belonging to a specific variable over time. However, values are plotted as lines and are presented on the same axis.
Best used for ... Comparing the trend of one value with another.
Share charts represent your data in relative terms. In share charts, values belonging to specific variables are divided into percentage shares, showing how each value contributes to the features of a variable over a date range.
Best used for ... Exploring how the features of a variable changed over time
Trends charts reveal how patterns in your data evolved over time. You need a date variable in your data to create trends charts. Start plotting trends by selecting a chart type in the Trends panel.
Date variables will be plotted on the x-axis of your chart. Quantitative variables will be plotted on the y-axis.
Switching between charts in your project's Trends panel means that you can easily compare different representations of your data over time.
Changing the type of chart won't reset the active filters you have applied. Use the chart dropdown in the top left to switch between types of charts.
After you've created trends charts you can make changes to the values and variables shown in them. You can change the quantitative variable shown on the y-axis as well as the variables and values shown in the charts.
In addition, you can add annotations to your charts or aggregate the data that is shown across different time periods.The quantitative y-axis variable determines how and what data is counted in your Trends charts. It can be aggregated across time periods. Variables are the fields in your data. Values are the data points within those fields.
To change the quantitative variable shown on the y-axis, select another option from the dropdown menu currently representing your y-axis variable. This sits at the top of the Trends panel.
Changing the y-axis variable changes what aspect of your data is used to count values.
Changing the variables that are shown in Trends charts changes what data fields you are visualising.
Changing the values that are shown in Trends charts changes what features of your variables are represented.
Adding annotations to your charts is a useful way to contextualize your analysis with significant events.
Annotations are placed at the bottom of x-axis and are marked by a vertical dashed line on your chart. Start adding annotations by clicking on the 'Add annotation' icon from the icon list at the top right of your project's Trends panel.
Annotations should be clear and short descriptions of significant events that affect trends in your data. Remove them by clicking on the description and selecting the trash icon below the text box.
You can aggregate the quantitative variable that is displayed on the y-axis of your trends chart. Aggregation operations like average, count, min and max will change the representation of the y-axis to summarise your data.
For instance, choosing count from the aggregation dropdown will change the y-axis so that it represents every instance of your variable. Averaging represents your variable or value as an average of the number of times that variable or value appears over a given date range.
Aggregating trends only changes how data is represented on the y-axis. The date range represented on the x-axis remains the same.
Alongside using the representations of your variables in the sidebar to filter your data, you can use the date histogram at the bottom of the Trends panel to represent only data appearing in a specific date range.
Use the date range histogram at the very bottom of your project's Trends panel to filter the dates presented in your chart. Clear date filters using the 'Clear' button in the right sidebar.
Click and drag your mouse over the date range histogram to filter your data by date. Drag the filter to the left or right to reposition it. Drag it's bars to resize it.
The Trends panel is a quick way to find interesting patterns in your data. In fact, they are so interesting that you might want to immediately share them on social media. You can either save trends charts as insights or export them to your computer.
Click on the export icon in the top left of the Trends panel to export the current Trends chart as an image.
Trends charts are often composed of separate elements which get downloaded independently. To export them as a group, save the chart as and download that.
When you save trends charts as insights you can add text, descriptions and statistics to supplement the chart. You can present trends charts directly inside of your project's Insights panel.
Capturing trends charts as insights then downloading that insight can be a useful way to collectively download all elements of a trends chart.
We know that data isn't always clean and simple.
Have a look through these topics if you can't see what you are looking for.