Skip to main content
Generating a Trend Radar

This article walks you through the steps for creating a Trend Radar.

Ö
Written by Öykü Aygül
Updated this week

The Trend Radar helps you identify and analyze emerging trends in your data. You can access it in the “Trend Radar” section of the “Analyze” view.

In the Trend Radar, each bubble represents a trend, with its size indicating its growth momentum. You have the flexibility to define what the radial and angular axes represent, and the AI will automatically estimate the corresponding values for each trend.

Angular scale: Select the metric for the angular axis, and the AI will evaluate each trend to assign a value. Bubbles will be positioned in their respective sectors (i.e. slices) based on these values, with higher values appearing further clockwise. For instance, if you enter 'Impact on Society,' trends with a greater societal impact will be located further clockwise on the chart."

Radial scale: Choose a metric for the radial axis (which extends outward from the center of the chart), and the AI will assign a value to each trend. For example, if you enter 'Trend time horizon', trends with a longer time horizon will be positioned further from the center along the radial axis.

Area of interest: Enter your area of interest to assist the AI in identifying trends related to that topic. If you are confident that all your text data is relevant, you may leave this field blank.

Language: Trend titles and summaries will be generated in the selected language, irrespective of the language used in the input dataset.

Approach: The radar can discover trends and trend areas using a bottom-up approach ( i.e. letting AI discover the underlying trends within the data) as shown in the example below. Alternatively, you can choose to manually add the trends and trend areas you are aware of. The latter gives you control over the trends and trend areas displayed on the radar.

Use predefined trend areas and trends

Once you choose to use predefined trends, you can add them using the “Add sector” button.

If you keep the “Discover additional sectors” setting enabled, in your radar, you will receive all specified trends and trend areas, along with additional ones identified through a bottom-up approach. If you disable this setting, your output will include only the sectors/trend areas you define. You can also adjust the discovery of trend setting for each trend area individually.

Fill in the values for sector and trend names. Ensure that these names are clear and easily understood by the AI. Instead of complete sentences, write trends as phrases of fewer than ten words. Lengthy text can make the radar appear cluttered.

Add more sectors to the radar through “Add sector” and remove the sectors or trends through the X symbol. Enabling “Discover additional trends” will identify more trends bottom-up, and disabling it will restrict the output to the trends you listed. Once you click “Save”, the radar will be generated accordingly.

Time period

Time period affects the values used for momentum calculation, which reflects bubble size. You have the options listed in the dropdown. To exemplify:

  • Choose "Annual" to compare momentum between the two most recent 12-month periods.

  • Choose "Monthly" to compare momentum between the last two months.

  • Other options are available based on the duration of your data.

Please note that the time periods you can select depend on the date range of your data set.

To download your radar, click the download icon located in the top left corner.

To publish your radar, use the icon to the right of the download icon. Ensure that publishing is enabled from the Insight Booster list before proceeding.

Did this answer your question?