Website scraping

Website scraping lets your receptionist learn from your existing website: FAQs, service descriptions, policies, and anything else published on your site.

Source types

Open Business knowledge from the Overview tab of your business page, then choose Website or Single page. If you have no sources yet, the options read Add your website and Add a single page.

URLs must use HTTPS.

How an entire website is read

Reception.ai reads your sitemap and homepage links, then selects the pages most likely to contain business information, such as services, pricing, team, locations, and help articles. Pages such as galleries, blog posts, and legal notices are skipped.

Booking platforms and social profiles, such as Facebook, Yelp, Vagaro, Booksy, or Calendly pages, are always read as a single page.

Reading the rest of your site

Onboarding runs a quick scan only, so it’s done in under a minute. To import the remaining pages, select Read the rest of your site on the Home page, or Read more pages on the website source. A deep scan can take several minutes.

Status

While processing, a source shows its current step, such as crawling or extracting. When done, it shows Ready or Error.

Select the page count on a website source, such as 12 pages, to see which pages were read.

If a website can’t be read, check the address. Some sites block automated access, for example with anti-bot protection. In that case, add individual pages or upload the content as a file.

Keeping content current

Websites change over time. Select Refresh on a website or page source to read it again. A website that has only had a quick scan shows Read more pages instead. If a source failed, select Try again. Refresh after changing pricing, services, or policies on your site.