Managing Content Sources for Your AI Assistant Chatbot

Your AI Assistant is only as good as the knowledge you give it. The Content tab lets you feed your chatbot with information from multiple sources — uploaded files, web pages, FlowMattic Tables, and WordPress posts. The more relevant content you add, the better your chatbot answers.

How It Works

When you add content, FlowMattic uploads it to your OpenAI Vector Store. The chatbot then searches this knowledge base every time a visitor asks a question, finding the most relevant information to craft its response.

Source 1: Upload Files

Upload documents directly to your chatbot’s knowledge base.

Supported File Types

You can upload PDFs, text files, markdown files, and other document formats supported by OpenAI.

How to Upload

1. Open the Content tab in your AI Assistant
2. Expand the Upload Files section
3. Click Add Source
4. Select your file and upload it
5. Wait for the status to show Completed

Managing Uploaded Files

Once you have files uploaded, you can:

• View All Files — Opens a file manager showing all uploaded documents with their status
• Search — Filter files by name in the file manager
• Delete — Select files and remove them from the knowledge base

Each file shows a status badge:

• Completed — File is processed and available to the chatbot
• In Progress — File is being processed
• Failed — Upload or processing failed — try again

Source 2: Crawl Web Pages

Add knowledge from any publicly accessible web page. This is useful for FAQ pages, help centers, product pages, or any content hosted outside WordPress.

How to Add URLs

1. Expand the Crawl Webpages section
2. Enter a URL in the input field (e.g., https://yoursite.com/help-center)
3. Click Add URL to add more URLs
4. Click Crawl All to start crawling all URLs at once

Tips

• Add your most important pages — FAQ, pricing, product descriptions, terms of service
• Each URL is crawled independently, so you can add as many as you need
• The crawled content is automatically uploaded to your Vector Store
• If a page changes, re-crawl it to update the knowledge base

Source 3: Connect FlowMattic Tables

If you store data in FlowMattic Tables (customer records, product catalogs, inventory, etc.), you can connect them directly to your chatbot.

How to Connect a Table

1. Expand the Connect Tables section
2. Select a table from the dropdown — it shows all your FlowMattic Tables with row counts
3. Configure Auto Sync to keep data up to date:

4. Click Sync Table Now to do the initial sync

Note: Auto sync has a 5-minute cooldown to prevent rapid repeated syncs when multiple rows change in quick succession.

How It Works

FlowMattic exports your table data and uploads it to the Vector Store in chunks (10,000 rows per chunk). The dropdown shows you how many chunks will be created based on your row count.

Source 4: Sync WordPress Posts

Connect your WordPress content directly — pages, posts, and any custom post type.

How to Sync Posts

1. Expand the Sync WordPress Posts section
2. Click Add Post Type
3. Select a post type from the dropdown (Pages, Posts, Products, or any registered custom post type)
4. Add more post types if needed
5. Click Sync Now to start syncing

What Gets Synced

FlowMattic exports the content of all published posts for each selected post type and uploads them to your Vector Store. This includes the post title, content, and key metadata.

Keeping Content Updated

After the initial sync, click Sync Now again whenever you publish new content or make significant updates. The status display shows when the last sync was completed.

Source 5: Google Docs & Notion (Coming Soon)

Direct integrations with Google Docs and Notion are currently in development. These will allow you to connect documents and databases from these platforms as content sources.

Best Practices

1. Start with your most-asked questions — Upload FAQ documents, help articles, and product guides first
2. Be specific — The more focused your content, the better the chatbot’s answers. A targeted help doc beats a generic brochure.
3. Keep it updated — Re-sync tables and re-crawl pages when your content changes
4. Test after adding content — Ask your chatbot questions about the newly added content to verify it’s working
5. Don’t overload — Quality matters more than quantity. Adding irrelevant content can confuse the chatbot.

What’s Next?

With your knowledge base set up, your chatbot can now answer questions intelligently. Next steps:

• Setting Up Lead Capture and Meeting Scheduling — Add interactive actions to your chatbot
• Customizing Your Chatbot Appearance — Make your chatbot match your brand