Scrunch home

Yes, Scrunch tracks sentiment about brands in AI responses and categorizes it as positive, mixed, or negative while showing how sentiment changes over time, which topics and personas are driving sentiment, and how brand sentiment compares to industry peers.

Additional context: Mixed sentiment indicates AI responses that include both pros and cons or neutral observations about a user’s brand.

Example

For example, a Scrunch user could track sentiment expressed toward their brand across multiple dimensions, including:

Cross-platform analysis

See the breakdown of sentiment for tracked AI responses across all AI platforms (ChatGPT, Gemini, AI Overviews, etc.).

Platform-specific analysis

See the breakdown of sentiment for tracked AI responses in specific AI platforms.

Filtered analysis

See the breakdown of sentiment for tracked AI responses across different personas, funnel stages, regions, and topics.

Prompt-level analysis

See sentiment expressed for specific prompts in different AI platforms (and view full AI responses).

Trend analysis

See how sentiment rises or falls over time.

Topic and persona analysis

See which topics and personas drive positive or negative shifts in sentiment.

Industry analysis

See how sentiment compares to brands in the same space.

Follow-up question: How does Scrunch determine if sentiment is positive, mixed, or negative?

Scrunch uses a machine learning model trained specifically on AI responses to analyze tone and positioning. Positive sentiment indicates language that is favorable or endorsing. Mixed sentiment indicates language that includes both pros and cons or neutral observations. Negative sentiment indicates language that highlights drawbacks or positions competitors as more desirable.