Skip to content
Knowledge and answers

Train the agent on your website

Knowledge is the set of pages, files and notes the agent searches before it answers. Point it at your website and it takes what a visitor could read; anything behind a login, a password or a cookie wall it cannot reach, and neither can your customers. The gaps you have to fill by hand are the things your website never said out loud.

Updated 2026-10-08 · 5 min read · More in knowledge and answers

What the crawl actually takes

It maps the site first — its sitemap and the pages your home page links to — and lets you pick the sections worth reading. Then it reads them, key pages and navigation first and every section after in proportion to its size, until the site or your plan's knowledge allowance runs out (5 MB on Free, about 500 pages; 50 MB on Growth; 250 MB on Scale). A small site takes a minute or two and a big one a few hours; its pages are searchable as they are read.

It takes what is public. A staging site behind HTTP auth, a customer portal behind a login, a PDF that only exists in a shared drive: none of those are reachable, and none of them will appear. If a page is important and the crawler missed it, add its address by hand rather than assuming it will turn up later: the crawl starts from that page, so it is read first.

Fill the gaps by hand

Three kinds of source can be added directly: a URL, a file, or a plain note you type. Notes are the underrated one. Most of what a support agent needs to know is not on any page — the hours your support line actually keeps, which carriers you ship with, what counts against a plan's limit this month, what happens to a return after thirty days.

Write a note the way you would answer the question out loud, in a sentence or two. Short, specific notes retrieve better than a long document where the useful sentence is buried in boilerplate.

  1. Find the questions

    Open Knowledge › Unanswered. Every question the agent found nothing for, said it did not know, or had its reply rated down is there, grouped with the other ways people asked it.

  2. Write the answer

    Click Write the answer on each one, in the words you would use on the phone. It is saved as a Q&A answer, which the agent gives ahead of your pages and files whenever someone asks that question or something close to it.

  3. Check it took

    A few minutes later the Answered list says how many of the ways it was asked now get your answer, and whether anyone has been missed since.

Keep facts out of the instructions

The agent's instructions are how it behaves: its tone, its house rules, what it must never promise. Its knowledge is what it believes to be true. The two get confused constantly, and the confusion is expensive in exactly one direction.

A price, a plan limit or a delivery time typed into the instructions becomes something the agent asserts with total confidence, forever, with no page behind it and no way for the crawl to correct it when you change your website. We have watched an agent quote a price that existed nowhere except a hand-typed line in its instructions — to a real customer, twice. Put facts in knowledge, where they have a source and can be updated.

Keep it fresh

Re-crawl after a meaningful change to your site — new pricing, a new returns policy, a new product line. Knowledge is a snapshot, not a live mirror, and a page you updated this morning is not updated in the agent until it has been read again.

Remove what has gone stale, too. An old promotions page that still ranks on your site is a page the agent can quote from, and a discount that expired in March is worse than no answer at all.

FAQ

Related questions

Enough to answer the questions people ask, and no more. A hundred well-chosen pages beat a thousand that include every blog post and archive page — noise competes with the passage you actually wanted.

Still stuck? A person will answer.