14 min
Lesson 3 of 4
AI Chatbots & Customer Engagement
From the course WordPress AI Plugins & Tools
Adding an AI chatbot to your WordPress site can dramatically improve customer engagement and reduce support workload.
Setting Up AI Engine Chatbot
AI Engine includes a built-in chatbot that you can train on your own content:
Configuration
- Go to AI Engine → Chatbot
- Set the chatbot name and greeting message
- Choose the AI model (GPT-4o-mini recommended for cost)
- Set the system prompt defining the bot's personality
- Configure appearance (colors, position, icon)
Training on Your Content
The chatbot can use your WordPress content as context:
- Go to AI Engine → Embeddings
- Select which post types to index (posts, pages, products)
- Click "Create Embeddings"
- The chatbot now answers questions using your content
This is essentially RAG (retrieval-augmented generation) built into WordPress.
Chatbot Best Practices
System Prompt Example
"You are a helpful assistant for [Your Business]. Answer questions about our products and services based on the provided context. If you don't know something, say so and suggest the user contact support at [email protected]. Be friendly, concise, and professional."
What Works Well
- FAQ-style questions about your products/services
- Navigation help (finding specific pages)
- Basic support queries
- Lead qualification questions
Limitations to Communicate
- Can't process orders or payments
- May not have the latest information
- Complex issues should go to human support
Measuring Impact
- Track chatbot conversations in AI Engine's logs
- Monitor common questions to improve your content
- Measure support ticket reduction after deployment
- A/B test chatbot placement and greeting messages
Key Takeaways
- •AI Engine's chatbot uses RAG to answer questions from your WordPress content
- •Train the chatbot by creating embeddings from your posts, pages, and products
- •Set clear system prompts defining personality and limitations
- •Track conversations and common questions to improve content and reduce support load
0 of 4 lessons marked complete
Progress is kept in this browser only.