What the chatbot knows — Gorilla Dash

The chatbot answers from your own content. On Knowledge Sources you choose which:

Source	What it contributes
Website Pages	Everything published on the attached websites.
Products & Services	Your active catalogue.
Locations	Each tribe's profile, address and contact details.
Case Studies	Your published Our Work posts.
Blog Articles	Your published articles.
Uploaded Documents	Files you add specifically for the chatbot.

Your internal content is never used

The Knowledge Base and Resource Library are deliberately excluded. Those hold staff material — price lists, operations manuals — and a public chatbot must not be able to quote them.

Uploading documents

Add PDF, Word (.docx), .rtf, .odt, .txt, .md, .csv and .html files, up to 20 MB each. A document can be scoped to a single tribe, which is how you give one tribe its own price list or service list without the others seeing it.

THE HIGHEST-VALUE UPLOAD

The questions your team answers by phone every week. Write them down once, upload the file, and the chatbot handles them for you.

It is only as good as your content

A thin website produces a chatbot that cannot help. If it keeps failing to answer something, the fix is nearly always to write the page rather than to change a setting.

Re-index after a big change

ENABLING A SOURCE DOES NOT MAKE IT SEARCHABLE IMMEDIATELY

Content has to be indexed before the chatbot can use it. Ordinary edits are picked up as you make them, but after switching a source on or loading a lot of new content, press Re-index and give it a few minutes.

Deleting a document

Removes it from the chatbot's knowledge straight away. Worth doing promptly when a price list goes out of date — an assistant quoting last year's prices is worse than one that says it does not know.