Add Websites

Train your chatbot with content from web pages.

How It Works

When you add a website URL, Leezy:

1. Fetches the web page content
2. Extracts text (removes navigation, ads, etc.)
3. Processes and embeds the content
4. Makes it available for chatbot responses

Adding a Website

1. Go to your chatbot's Sources page
2. Click Add Source
3. Select Website
4. Enter the full URL (including https://)
5. Click Add

Example URLs:

• https://example.com/help/getting-started
• https://docs.example.com/faq
• https://example.com/products/feature-guide

What Gets Crawled

Single Page Mode

By default, only the specified URL is crawled. This is useful for:

• Individual help articles
• Specific FAQ pages
• Product pages

Content Extraction

The crawler extracts:

• Main content text
• Headings and paragraphs
• Lists and tables
• Code blocks

The crawler ignores:

• Navigation menus
• Sidebars
• Footer content
• Advertisements
• Scripts and styles

Best Practices

Choose Good URLs

Ideal pages for training:

• FAQ pages
• Help center articles
• Documentation pages
• Product descriptions
• Blog posts with useful information

Avoid These URLs

• Login-required pages (authentication blocks crawling)
• Dynamic single-page apps (content may not load)
• Pages with mostly images/videos
• Very long pages (may hit content limits)

Keep Content Fresh

Website content changes over time. To update:

1. Delete the old website source
2. Re-add the URL

Regular Updates

If your website content changes frequently, schedule regular re-crawls by removing and re-adding URLs.

URL Requirements

• Must include protocol (https:// or http://)
• Must be publicly accessible
• Cannot require authentication
• Must return HTML content

Processing Time

Web pages typically process faster than documents:

• Simple pages: Under 1 minute
• Content-heavy pages: 1-2 minutes

Troubleshooting

Crawl Failed

Common causes:

• URL is incorrect or page doesn't exist
• Page requires login
• Website blocks crawlers
• Server timeout

Solutions:

• Verify the URL opens in your browser
• Check if the page requires authentication
• Try a different page from the same site

Little or No Content Extracted

Common causes:

• Page content loads via JavaScript
• Content is mostly images
• Anti-bot protection

Solutions:

• Try pages with static HTML content
• Use documentation pages instead of marketing pages
• Consider uploading content as a document instead

Content Is Outdated

Website sources capture content at the time of crawling. To refresh:

1. Delete the existing website source
2. Add the URL again

Limits

Plan	Websites
Free	5
Starter	20
Pro	100
Business	Unlimited

Next Steps

• Create Q&A Pairs - Add specific answers
• Upload Documents - Add file-based content
• Test Your Chatbot - Verify training