The Problem
Traditional support channels create friction:- Live chat requires staffing and has limited hours
- Ticket systems are slow and impersonal
- Knowledge bases require users to search for answers
- Users abandon when they can’t get immediate help
The Solution
Deploy the BoostGPT chat widget as an always-on AI support agent that answers instantly, 24/7, trained on your product knowledge.Implementation
Step 1: Train Your Agent
Before deploying the widget, train your agent on your support content:- Go to Memory in your agent dashboard
- Upload your knowledge base:
- Product documentation
- FAQ pages
- Pricing information
- Policies (shipping, returns, refunds)
- Common troubleshooting guides
Step 2: Configure Agent Instructions
Set up your agent’s system instruction to handle support conversations:Step 3: Install the Widget
Add one line to your website:Step 4: Configure Greeting Messages
Set up welcoming greetings in Branding → Content → Message Greetings:Step 5: Enable SSO (Recommended)
If your site has user authentication, connect SSO to identify users:- Agent greets users by name
- Conversation history persists across sessions
- Support team can identify who needs help
Step 6: Add Proactive Triggers
Combine the widget with triggers to proactively offer help at friction points:Widget vs Triggers — When to Use Each
| Approach | Best For | Example |
|---|---|---|
| Widget (always visible) | General support availability | ”Chat with us” bubble on every page |
| Triggers (proactive) | Specific friction points | ”Stuck on checkout?” when user hesitates |
| Both (recommended) | Maximum coverage | Always-on support + proactive help |
Measuring Success
Track these metrics to evaluate your support widget:| Metric | Target | How to Measure |
|---|---|---|
| Resolution rate | 80%+ | Conversations resolved without human handoff |
| Response accuracy | 90%+ | Spot-check agent answers weekly |
| User satisfaction | 4/5+ | Add rating prompt after conversations |
| Support ticket reduction | 30-50% decrease | Compare ticket volume before/after |
| First response time | Under 5 seconds | AI responds instantly |
Optimization Tips
Review conversation logs weekly
Review conversation logs weekly
Check the Conversations tab to find:
- Questions the agent couldn’t answer (gaps in knowledge)
- Incorrect or confusing responses (needs instruction refinement)
- Common topics (may need dedicated documentation)
Expand knowledge base incrementally
Expand knowledge base incrementally
Start small, then add content based on real conversations:
- Launch with core FAQ content
- Review unanswered questions after 1 week
- Add missing content to memory
- Repeat monthly
Use web access for dynamic content
Use web access for dynamic content
Enable web access so your agent can fetch live data:
- Current status page information
- Latest pricing from your website
- Product changelog and release notes
Set up escalation paths
Set up escalation paths
Not everything should be handled by AI. Configure clear escalation:
- Billing disputes → “Please email billing@yourcompany.com”
- Account security → “Contact us at security@yourcompany.com”
- Complex technical issues → “I’ll connect you with our engineering team”
Next Steps
Widget Configuration
Customize appearance and behavior
SSO Setup
Identify your users automatically
Event Triggers
Add proactive messages
Best Practices
Tips for effective deployment