Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.shareofmodel.ai/llms.txt

Use this file to discover all available pages before exploring further.

In generative search, being mentioned is not enough. LLMs synthesise multiple sources into a single answer, which can result in mixed narratives — a brand may be praised by one source and criticised by another, sometimes within the same prompt. This effect is especially important at lower-funnel queries, where tone directly influences trust and decision-making. The Sentiment Score captures how a brand is portrayed within:
  • LLM-generated answers,
  • the sources used to construct those answers.
Each detected mention is assigned a sentiment score ranging from -1 to +1, reflecting how positively or negatively the brand is expressed. The score also incorporates the intensity of the sentiment — stronger opinions weigh more than neutral or marginal mentions.

​
How the score is calculated

The score is built from individual brand mentions. For each prompt:
  • mentions detected in sources and answers are analysed individually,
  • each mention receives a sentiment score from -1 to +1 using a standardised NLP model (Google Natural Language Processing where supported, with fallback methods otherwise),
  • the score reflects both direction (positive, neutral, negative) and intensity (magnitude weighting strongly expressed opinions more heavily).
Scores are aggregated across multiple levels (prompt, source, topic, period), always starting from the individual-mention level to preserve context and avoid bias in higher-level analysis. The aggregated score also ranges from -1 to +1:
  • negative sentiment is subtracted from positive sentiment,
  • neutral mentions contribute to a balanced score around zero.

​
How to read the score

Read the Sentiment Score as a directional signal, not an absolute judgment.
  • A positive score suggests the brand is generally framed favourably.
  • A neutral score often reflects factual or descriptive mentions.
  • A negative score indicates recurring critical or unfavourable framing.
Sentiment is context-dependent — the same brand can be framed differently depending on the prompt, source, engine or stage of the user journey. The score is most meaningful when analysed across prompts, topics, sources and time periods.

​
How this KPI fits with other signals

Read alongside Presence Rate and Visibility Score to understand not only where a brand appears, but how prominently and how it is framed in generative answers.

​
What’s next

Sources & Links

See where mentions originate from.

Fan-out Queries

Understand the internal queries behind the answer.