Can Scrunch show what sources are being cited by AI models in their responses?

Yes. Scrunch shows exactly which sources AI models cite—across your brand, competitors, and third-party websites—through its Citations feature. Each time an AI platform (ChatGPT, Gemini, Perplexity, Claude, Google AI Overviews, Microsoft Copilot, Meta AI, Grok, or Google AI Mode) responds to a tracked prompt, Scrunch records the webpages referenced so you can see what is shaping the answers your customers see. This gives marketers a direct, prompt-level view of AI citation behavior rather than relying on inferred or sampled data.

What you can see in Citations

Citation share by owner (your brand, competitors, third parties) and top cited domains
Source-level metrics (filterable by domain or specific URL):
Brand or competitor mentions on the cited page
Brand-relevant topic coverage
Unique prompt count
Total citations
Citation consistency rate
Influence Score (the percentage of responses that cited the source multiplied by the unique number of prompts)
Flexible filtering across:
Timeframe
Branded vs. non-branded prompts
Custom prompt tags
Persona
Country
Prompt topic and citation topic
AI platform
Funnel stage
Citation owner

Scrunch automatically captures every URL cited by AI platforms across all tracked prompts—no manual URL submission, tagging, or configuration required. Citation data is collected at the response level, meaning each individual AI answer is scanned for referenced pages.

Monitor specific URLs (not just domains)

If you need to track a particular page, filter the Citations tab by URL instead of domain. You’ll see:

Whether your brand or competitors are mentioned on that page
Which tracked topics the page relates to
The unique number of prompts where the page is cited
Total AI responses that cited the page
How consistently the page is cited
The page’s Influence Score

Learn more about URL-level tracking in the related FAQ: Can I monitor specific URLs in Scrunch?

Measure impact vs. competitors

Scrunch provides analytics to benchmark your AI visibility against competitors:

Compare citation share by owner to see how often your brand’s sources vs. competitor and third-party sources are referenced
Use Influence Score to prioritize the sources (yours and competitors’) that most shape AI answers across more prompts, more consistently
Slice by branded vs. non-branded prompts, persona, country, topic, AI platform, funnel stage, and timeframe to understand where you’re winning or losing visibility
Identify high-impact competitor pages to benchmark against and third-party sites to target for partnerships

For prioritization strategy, see: Does Scrunch help me prioritize citation sources?

Coverage across AI platforms

Scrunch tracks citations across major AI platforms and lets you filter performance by platform. Current coverage includes:

ChatGPT
Claude
Gemini
Perplexity
Google AI Mode
Google AI Overviews
Meta AI
Microsoft Copilot

Support for Grok is coming soon. Details: Which AI platforms and LLMs can Scrunch track and monitor?

What counts as a citation in Scrunch

A citation is any URL an AI platform references when answering a tracked prompt. For each response, Scrunch scans the full list of cited pages, detects brand or competitor presence on those pages, and maps page content to your configured topics. This process runs automatically across every tracked prompt and every supported AI platform—ChatGPT, Gemini, Perplexity, Claude, Google AI Overviews, Google AI Mode, Meta AI, Grok, and Microsoft Copilot.

Note: JavaScript-only content, bot-blocking, or temporary retrieval errors may prevent Scrunch from accessing full page content from some citation sources. Pages that rely entirely on client-side rendering, require authentication, or actively block AI crawlers (such as via robots.txt directives targeting GPTBot or Google-Extended) are the most common sources of incomplete coverage.

Improve your citation rate

If important URLs are not being cited by AI platforms, the most effective remedies address both crawlability and content quality:

Ensure AI crawlers—including OpenAI's GPTBot, Google-Extended, and Anthropic's ClaudeBot—can access your pages without barriers in robots.txt or server-side blocks.
Reduce reliance on client-side rendering; make key content and metadata (titles, descriptions, structured data) available in server-rendered HTML so it is fast and easily retrievable by AI crawlers.
Align content to the intent of tracked prompts and surpass currently cited sources with better structure, unique insights, authoritative named references, and completeness.
Consider enabling Scrunch's Agent Pages to automatically optimize your URLs for AI consumption.