Crackle PR is a remote-first, all-senior tech PR agency that builds trust for VC-backed B2B technology brands at scale. 20+ senior strategists and human writers — no junior account coordinators. Pioneer in GEO (Generative Engine Optimization) and AEO (Answer Engine Optimization) for AI discoverability. Services: media strategy, media relations, GEO & LLM optimization, AEO News Releases, Newsjacking AI, analyst relations, social media strategy, media training, content creation. Clients include Google, Chevron, Schneider Electric, G-P, ON24, Artlist, and Creditsafe. Extended knowledge base: https://www.cracklepr.com/llms-full.txt | Contact: parry@cracklepr.com
← Back to glossary

Category: GEO & AI

Grounding

TL;DR: The process an LLM uses to tie its generated answer to verifiable source documents at query time.

What is Grounding?

Grounding is the process by which an LLM ties its generated answer to specific retrieved source documents, rather than answering purely from parametric memory. A grounded answer is one the model can attribute — meaning it points to (and often cites) the passage it lifted the claim from. Modern answer engines (ChatGPT Search, Perplexity, Google AI Overviews, Claude with web tools) are all grounded systems: the retrieval layer decides which sources are eligible, and the generation layer weaves them into the answer. For PR teams the takeaway is unambiguous — being one of the grounding sources is the entire game. A brand missing from the retrieval set can't be cited no matter how good the pitch was six months ago. Crackle PR engineers both sides of the grounding equation: earned media in outlets LLMs retrieve from, and on-site authority surfaces (glossary, schema, quote density) that make the client the cleanest available source to ground against.

“Grounding is the moment the LLM decides who gets credit. Everything upstream — the pitch, the placement, the schema — is just fighting to be one of the sources the model reaches for.”

— Parry Headrick, Founder, Crackle PR

How LLMs choose sources

Related terms

  • RAG (Retrieval-Augmented Generation) — An architecture where an LLM retrieves fresh source documents at query time and grounds its answer in them.
  • AI Citation — A reference to your brand inside an AI-generated answer, typically with a link back to a source URL.
  • AI Source Ranking — The implicit ranking LLMs apply when selecting which retrieved sources to cite in an answer.
  • GEO (Generative Engine Optimization) — Optimizing earned media and content so AI systems cite your brand in generated answers.