Does Scrunch support international prompt tracking and multiple languages?
Yes. Scrunch supports prompt tracking in every country and records the exact AI responses and citations for prompts written in any language. For AI platforms that support geolocated search—such as ChatGPT, Google AI Overviews, and Perplexity—Scrunch returns country-specific results. Scrunch currently serves customers across the US, Europe, and Japan.
How it works
Set a primary target country in the AI Context tab. This becomes the default geography for all new prompts.
Create market-specific personas with country and language settings, then assign prompts written in the local language to those personas.
Choose which AI platforms to track—one, several, or all supported at once. Scrunch supports ChatGPT, Claude, Gemini, Perplexity, Google AI Mode, Google AI Overviews, Microsoft Copilot, and Meta AI.
Compare responses and citations by country and platform. Filter results by specific LLMs to understand how answers vary across markets and models.
Example
A global project management software company wants to track brand presence across North America, Europe, and Asia. The team sets the US as the primary country in AI Context. For Germany, they create a “German Enterprise Buyer” persona, set the country to Germany, and add German-language prompts like “Welche ist die beste Projektmanagement-Software?”. They then compare how AI platforms answer “What is the best project management software?” in the US versus Germany, including differences in competitors mentioned and sources cited. The same approach can be repeated for other countries and languages.
If responses don’t vary by country
Language mismatch: AI favors sources in the same language as the prompt. Using English prompts for non-English markets can return similar results to English-speaking markets.
Limited local content: AI may default to English sources if stronger local-language content isn’t available.
Broad targeting: Add specific cities or regions to prompts (e.g., “in Madrid,” “near Toronto”) to surface more localized results.