Hi @grazianodelucaa2a!  When creating a Live chat experience it's all about finding the right balance between automation and human support. But always keeping the Customer Experience as your top priority!

Here are some of the key things I’ve learned along the way:

✅ Make the Agent Experience Smooth

• Custom Smart Views: Create role-specific views (e.g., Billing Inquiries, Outage Reports, High Priority) using case properties and keywords to streamline triage.  

• Case Routing Rules: Leverage the Rule Engine to assign cases based on agent skill, language, or queue load—minimizing resolution time.  

• Context Cards: Integrate CRM and billing data (e.g., last bill amount, open service orders) for quick, informed responses.  

• Canned responses are good for common questions (like “Did my payment go through?”), but be sure that when creating these your agents use their own words wen elaborating.  Here's a guide to best practices for agent engagement. 

---

📊 Keep an Eye on the Right Metrics

• First Response Time (FRT)  

• Average Resolution Time  

• Chatbot Containment Rate (bot-only resolution)  

• CSAT by Channel/Agent  

• Top Intent Categories (from bot/live chat)  

• Live Chat Abandonment Rate  (this is a key indicator to understanding if your customer is having a positive experience) 

💡Dashboard Tips:  

• Use real-time widgets to monitor queue health and agent activity.  

• Build intent heatmaps to identify automation opportunities.  

• Segment reports by issue type (e.g., billing, outages, account access) to uncover trends and optimize workflows.  

🤖 Build Smart, Helpful Bots

• Start with top intents based on historical data (e.g., “Report an Outage,” “View My Bill,” “Start/Stop Service”).  

• Use quick replies and buttons to reduce user effort.  

• Always offer a “Speak with an Agent” fallback after a few failed attempts.  

🔀 Workflow & Rule Engine Setup:  

• Follow a Bot → Workflow → Case Creation sequence:  

• Bot collects key info (e.g., address, account number, issue type)  

• Workflow validates and enriches data  

• Rule Engine creates and routes the case  

• Tag and classify interactions early for better analytics and routing.  

• Test human handoff rules thoroughly to ensure smooth transitions when needed.  

⛓️ Configuration Sequence:  

• Define intents and design dialogue trees in Conversation Designer  

• Set up workflows for each bot action (e.g., fetch account info, create case)  

• Build rules for escalation, tagging, and case creation  

• Configure dashboards and agent views  

• Test the full bot → case → agent → resolution journey  
  

---

🛠️ Pro Tips

• Remember bots are only as smart as the people who build them 😉

• Always consider the customer experience first when build your chat bot, the smarter you build your bot, the less often an agent has to take over. 

• When creating your human handoff flows think about your customer needs and determine when Synchronous vs Asynchronous chat works best. You don’t want customers stuck in bot limbo.

• Build everything in layers: intents → workflows → rules → dashboards → agent views.

• Tag and classify early in your process—it helps with routing and reporting later.

---

📚 Want to Learn More?

• Check out the Sprinklr Help Center for how-tos.

• Sprinklr University has great training modules. Conversational AI Pro has several great session to help get you more familiar with AI capabilitues and advanced training.

• Read our Blog: Best practices for building efficient chatbots

💖 I would love to hear from some of our customers, on more tips and tricks!