ChatGPT is the most widely used generative AI platform in the world. The platform you can least avoid — and one whose mechanics are more complex than they appear at first glance.
How does ChatGPT work?
ChatGPT operates in two strategically different modes. In the default mode, it generates answers based on training data: what the model has learned up to its cutoff date. In search mode, ChatGPT retrieves live web pages through a connected index and can show sources alongside the answer. The user determines which mode is active, often without realising it.
What has the most impact potential
- Long-term, broad web presence — training data is the foundation of the default mode
- Brand frequency: the more often your brand is correctly described across the web, the stronger the training representation
- Indexing in the search engine that ChatGPT queries for live retrieval
- Off-site presence: external sources describing you reinforce the training data
- Clear, citable phrasings in case search mode is active
Considerations
- ChatGPT rarely shows explicit source links in default mode — brand mention without a clickable source is the goal here
- In search mode, sources are visible; observability therefore varies strongly per mode
- Factual inaccuracies are difficult to correct — prevention through consistent presence is more robust than reactive correction
- Track your presence in the search index ChatGPT uses for live retrieval
Related in the hub
→ Want a comparison of tactics across platforms? Read 5.5 — Platform tactic overview.