To understand how Heartbeat works, here is an explanation of the key terms used across the platform.

Messages %
​

The y-axis plots the proportion of analyzed messages containing the linguistic features grouped into each theme. A message refers to a singular survey open-ended response, tweet or similar.

In the example above, we can see that in February, the theme 'Affordability' was mentioned in 4.2% of all analyzed tweets.

Daily, weekly, monthly, quarterly and annual views of your data

You can toggle between annual, quarterly monthly, weekly and daily views of your data using the options located beneath the chart.

Click on a point on the chart to reveal the verbatim examples used in the analyzed text and explore the quotes that contribute to the conversation. You can click on each quote to see it in full.

This helps to understand the context in which specific words or phrases are being used to provide a deeper insight into how the public discourse evolves over time.

When plotting your Custom Themes as a timeseries, you can add an overlay to plot metadata values over time. This allows you to correlate changes in your metadata with language changes.

To add this overlay, hit the Plot button and select the metadata key you want to plot. Please note: Only numerical metadata values are supported.

As well as a time series chart, Heartbeat allows you to plot your Custom Themes in other ways.

For example, you can see the Custom Theme breakdown against your metadata using the Categorical plot.

To switch to this view, select Plot, then Categorical, and select your metadata key from the drop-down.