03 / COMPARE
One brand. Three AI perspectives.
Transparion Pro Scan examines ChatGPT, Claude and Gemini using the same configured questions. This reveals similarities and differences in brand mentions, answers and sources.
SUPPORTED SYSTEMS
What Transparion checks in each system.
The model overview explains how each system is accessed. It does not show live answers or rate any system as generally better or worse.
Select a model to explore how it is analysed.
HOW IT IS ACCESSED
ChatGPT with web search
Transparion examines answers through the OpenAI API with web search. Visible sources and brand mentions are evaluated alongside each answer.
Included in Transparion Insight and Pro Scan.
Provider documentationANALYSIS SCOPE
Start with ChatGPT. Explore further with three systems.
Choose the scope that suits your question. The free analysis provides an initial view; Pro Scan offers a configurable comparison.
Scroll sideways to compare.
| Analysis scope | Transparion Insight | Transparion Pro Scan |
|---|---|---|
| AI systems | ChatGPT | ChatGPT, Gemini and Claude |
| Purchase-oriented questions | Automatically derived from the website | 10–30 visible, editable questions and your choice of target market |
| Competitors | Automatically identified vendors | Up to four selected competitors |
| Content & technology | One optional content page with AEO feedback | Up to five content pages, detailed briefs and technical SEO |
| Results & history | Online results through an email link | Individual model answers, account history, PDF and editable PowerPoint |
Perplexity is currently included in neither scan. This measurement does not cover other AI systems or personalized chat interfaces.
COMPARE FAIRLY
Consistent conditions make differences easier to understand.
Transparion uses model APIs and configured search tools. The results cover selected questions and points in time, not every conversation in the user interfaces.
- Keep questions and market consistent
- Compare the same prompt set, target region and competitors wherever possible. A new question changes the sample being examined.
- Read answers alongside their sources
- A brand mention and a link to your domain are separate signals. Check what the source actually supports and how your brand is described.
- Account for failures and model changes
- Failed answers do not count as zero visibility. Missing historical model details remain marked as unavailable. Changes to models or sources can affect the comparison.
- Understand the overall score
- The overall score uses all successfully evaluated answers. It is not a simple average of the rounded model scores; when answer counts differ, the systems contribute in proportion to those counts.
ILLUSTRATIVE OVERALL CALCULATION
ChatGPT: 6 out of 10, Claude: 4 out of 8, Gemini: 5 out of 10. Together, that is 15 mentions across 28 successful answers: approximately 53.6% AI Visibility.
EXPLORE IN MORE DETAIL
A shared foundation for your next analysis.
Save relevant prompt sets and compare follow-up analyses of the same website. In your Pro Scan result, you can open the actual answers from each model.
Use your free account to organize Pro Scans, website histories and scheduled analyses. You can track development after two completed Pro Scans of the same website.
19 € first individual scan · 14 € each additional individual scan
Configure Pro Scan Pro Scan features and pricing