Skip to main content

Sentiment Over Time and Narrative drivers

Understand how AI models describe your brand and whether the tone is positive, neutral, or negative

Written by Urszula Kucińska

It's not just about whether AI models mention your brand. It matters how they talk about it. Are they presenting it in a positive light, staying neutral, or pointing out drawbacks? The Sentiment Over Time chart and Narrative Drivers section in the Context tab help you answer these questions

Sentiment Over Time

At the top of the Context tab, the Sentiment Over Time chart shows the proportion of positive, neutral, and negative mentions of your brand over time. You can view the data in Daily, Weekly, or Monthly intervals.

By default, the chart aggregates sentiment across all AI models. If you filter the view to a specific model using the dropdown in the top right corner, it will show how that particular model's tone toward your brand has been changing.

You can also switch to the By AI model view using the toggle in the chart. This breaks down sentiment per model, so you can quickly see which models describe your brand more positively and which lean neutral or negative.

Narrative Drivers

Next to the Sentiment Over Time chart you'll find the Narrative Drivers section. While the chart above tells you the overall tone, Narrative Drivers show you the specific contexts behind that tone. It lists the topics and phrases that AI models associate with your brand, grouped by sentiment.

For example, positive drivers might include phrases like "real-time alerts," "cost-effective," or "sentiment analysis," meaning that when AI models speak positively about your brand, these are the features and qualities they highlight. Neutral drivers show the contexts where your brand is mentioned factually, without a strong positive or negative angle. If negative drivers appear, they reveal what aspects AI models frame critically.

The percentage bar at the bottom gives you a quick overview of the overall sentiment split.

For more on how AI models perceive your brand, see our article on Topics Analysis. If you have any questions, don't hesitate to contact us.

Did this answer your question?