Overview
This page explains how AI Brand Monitor collects data from AI platforms. Understanding this helps you get the most out of the platform.Data Collection Method
We Simulate Real User Experiences
Common question: âDo you use ChatGPTâs API to get responses?âAnswer: No. We simulate real user experiences in a browser - just like your customers do.
| Method | How It Works | Result |
|---|---|---|
| API Approach | Send request to API endpoint | API-specific response |
| Our Approach | Simulate user in browser | Real user experience |
Why This Matters
AI platforms behave differently between their API and consumer interface:Different Responses
ChatGPT.com may cite sources while the API doesnât
Different Features
Web search, real-time data, and citations vary
Authentic Results
We see exactly what your customers see
The Process
Hereâs what happens for each prompt you create:This happens for every prompt across every platform you have enabled -
every single day.
Data Classification
Intelligent Analysis
AI Brand Monitor uses AI (by tedi-2.1-reasoning-alpha-22022026) to classify and analyze all the data it collects:| What We Classify | How It Works |
|---|---|
| Domain types | Is it Editorial, UGC, Corporate? AI determines this |
| URL types | Article, Comparison, Product Page? Automatically classified |
| Sentiment | How positive or negative? AI evaluates the context |
| Brand detection | Which brands are mentioned? Scanned automatically |
| Source relevance | How important is this source? AI determines importance |
| Action suggestions | What opportunities exist? AI identifies gaps |
Processing Schedule
Daily Updates
What happens overnight?
What happens overnight?
- Your prompts are sent to all selected AI platforms (See your project settings)
- Responses are collected and stored
- Brand mentions are detected
- Sources are cataloged and classified
- Scores are calculated
- Actions are generated Fresh data is ready when you log in each morning.
Why overnight?
Why overnight?
We run prompts intelligently across platforms to:
- Ensure high-quality results
- Track how responses evolve over time
Summary
| Aspect | How It Works |
|---|---|
| Data collection | Real Browser simulation (not APIs) - authentic user experience |
| Processing | Daily updates with fair queue-based distribution |
| Classification | Tedi-powered analysis (not hardcoded rules) |
Frequently Asked Questions
How fresh is the data?
How fresh is the data?
Prompts run daily. You see yesterdayâs AI responses today - giving you the
most current snapshot of how AI platforms talk about your brand.
What if an AI platform is down?
What if an AI platform is down?
We automatically retry failed requests. If a platform is
temporarily unavailable, that specific chat is logged as failed but doesnât
affect your other data.
Do you cache responses?
Do you cache responses?
No. Each prompt gets a fresh response daily. AI platforms update constantly,
so caching would give you stale data.
Can I run prompts on-demand?
Can I run prompts on-demand?
Currently, prompts run on a daily schedule. But you can run test prompt on our Ai search tool This ensures consistent,
reliable data and fair resource distribution across all users.
Why browser simulation instead of APIs?
Why browser simulation instead of APIs?
API responses differ from what real users see. We prioritize accuracy -
showing you exactly what your customers experience when they ask AI about
your industry.
What if a platform has a captcha or blocks access?
What if a platform has a captcha or blocks access?
Our infrastructure is designed to handle this seamlessly. Just like a human, Tedi operates a real browser and can solve captchas, rotate proxies, and adapt to blocking measures. If a platform becomes temporarily inaccessible, we log that specific chat as failed, but it doesnât affect your other data. Within minutes or hours (not days), we automatically retry to retrieve the data.