What it means
Grounding is the practice of constraining an AI model's response to factual, attributable sources. A grounded answer comes with citations - links, references, or quoted passages that show where each claim originated. An ungrounded answer is the model writing from its general training, which can drift into hallucination.
Modern AI search engines like Perplexity, Google AI Overviews, and ChatGPT search are built on grounding. The engine searches the web, finds relevant pages, and asks the LLM to answer using only those pages as context.
Why it matters
For your business, grounding is what determines whether AI engines cite you or generate vague, inaccurate descriptions. The more easily an AI engine can find authoritative pages on your site that answer the user's question, the more likely it is to ground its answer in your content and cite you as the source.
How it's used
To make your content groundable:
- Write self-contained paragraphs with one clear claim each
- Include the specific facts the AI needs (location, service, price range, qualifications)
- Use schema markup to give AI explicit structured facts
- Make sure technical pages can be crawled by GPTBot, ClaudeBot, and PerplexityBot
- Build topical authority so multiple pages on your site reinforce the same facts
For Perplexity-specific grounding tactics, see our Perplexity SEO guide.