wordpress already makes a sitemap for you. retune reads it, you leave out the pages you don't want, and the bot answers from the rest. the chat goes on your site as one script tag.
it's free if you run your own copy of retune. that means retune runs on your own computer, with an agent that builds what you ask for. email me for that, or sign up and start on retune today.
the shop in the video is maple street bike shop. i made it up to show the feature. its site isn't wordpress, but the steps are the same.
reading a whole sitemap isn't switched on for new accounts by itself. sign up, then email me at olive@reframe.so and say you want the sitemap crawl on your bot. once it's on, a "from a sitemap" section shows up in the train tab and these are the steps.
/tag/**, /author/**. the ** means everything under that path. add /category/** too if you don't want the category lists.</body> in footer.php, under appearance, theme file editor.the chat bubble shows on every page of your site. a visitor asks in their own words, and the bot answers from the posts and pages it read.
if nothing on your site covers a question, the bot should say it doesn't know instead of guessing. it can still get one wrong now and then, so ask it a few real questions before it goes live.
if the default bubble looks wrong on your theme, you can make the chatbot look like your website. the embed code stays the same. and if you'd rather have the leads from the chat in a sheet, you can send chatbot leads to zapier.
run discover again. the pages it already read show as already trained, and only the new post is ticked. crawl that one.
discover only picks up new pages. if you edit an old post, the bot keeps the old text until it reads that page again. the sync tab does that on a schedule, and it's in keep your chatbot updated with website changes.
it reads the html your site sends. most themes put the post text right there, so that's fine. a page built with a lot of javascript, like tabs from a page builder, can come out thin. open a few rows after the crawl and ask the bot about them.
it only reads pages on your own domain. pdfs and images in the sitemap get skipped, and so do pages behind a login, like a members area.
it doesn't follow the "disallow" lines in robots.txt. if there's a page you don't want read, exclude it or untick it before you hit crawl.
one crawl takes up to 500 pages. if your site has more, crawl again. the pages already read show as already trained, so the next crawl takes the rest.
a theme update can overwrite footer.php. if you paste the code there, do it in a child theme, so it survives the update.
what it stores is the text of your public pages, in your bot in retune. it doesn't change anything on your wordpress site.
olive