CitePeak
Sign inGet started

Methodology

CitePeak queries AI answer engines via official APIs for the prompts you track, then analyzes each answer for brand mention, position, sentiment, and cited sources.

The AI Visibility Score (0–100)

Score = 100 × ( 0.40·Mention% + 0.30·SOV + 0.20·Citation% + 0.10·Sentiment )

  • Mention% — share of answers that mention your brand.
  • SOV — share of voice: your mentions ÷ (yours + competitors').
  • Citation% — share of answers citing your domain as a source.
  • Sentiment — tone (positive/neutral/negative), normalized to 0–1.

Engines & data sources

We use official APIs only (Perplexity Sonar, Anthropic Claude, Google Gemini, OpenAI Responses). Google AI Overviews and Copilot have no official monitoring API and are covered — when enabled — only via licensed SERP providers, clearly labeled.

ChatGPTPerplexityGoogle AI OverviewsGeminiClaudeMicrosoft Copilot
AI answers are non-deterministic. Scores may vary ±5–10% week to week as a natural result of model behavior. Low-confidence extractions are flagged for review. Reports are informational only and are not an endorsement or ranking.