Skip to main content
Brand Sentiment shows you the emotional and qualitative layer of your AI presence. Not just whether you appear, but how you’re described when you do. It surfaces the actual AI responses that contain your brand, breaks them down by attribute, and flags the characterizations that are helping or hurting you.

​
Sentiment scores

The top of the page shows your overall sentiment distribution — the share of mentions that are positive, neutral, and negative — compared to your previous period and against your tracked competitors. Sentiment is calculated per AI call: each response that mentions your brand is classified by tone, then aggregated. A single AI answer can contain both positive and negative signals about different attributes, and these are counted separately.

​
Attribute-level breakdown

Below the headline score, sentiment is broken out by specific attributes AI engines associate with your brand:
  • Ease of use — how AI characterizes the effort required to get value from your product
  • Value — how AI frames your pricing and ROI relative to alternatives
  • Reliability — how AI describes your consistency and trustworthiness
  • Support — how AI characterizes help availability and responsiveness
  • Other attributes surface automatically based on what the scan data finds
Green attributes are where AI is representing you favorably. Red or mixed attributes are where AI has absorbed negative signals — often from competitor-written comparison content, old reviews, or a single high-authority page that frames you unfavorably.

​
Actual AI response examples

The most useful part of Brand Sentiment is the AI Calls section: real, unedited AI responses where your brand was mentioned. For each response you can see:
  • Which engine generated it
  • Which prompt triggered it
  • The full text of the AI’s answer, with your brand mention highlighted
  • The sentiment classification for that specific call
Reading these directly tells you more than any aggregate score. You’ll often find that a single commonly cited article or a specific competitor’s comparison page is responsible for a consistent negative framing.

​
What to do with negative sentiment

Negative sentiment in a specific attribute usually points to a content fix:
  • Negative on “value” — create a detailed pricing rationale page and a ROI calculator that AI can cite as the authoritative answer
  • Negative on “ease of use” — publish a concrete onboarding guide with measurable time-to-value claims
  • Negative on “reliability” — surface case studies and uptime data in formats AI can extract cleanly
Take the attribute with the worst score, find the specific AI responses driving it, trace the source, and counter it with better content in Brand Studio. Sentiment shifts typically appear within 30–60 days of publishing a strong counter-piece.