Overview
Helena’s Sentiment Insights feature helps MSP technicians understand how customer users are feeling and what they’re trying to achieve — before replying to a ticket.
By analyzing tone, language, and message patterns, Helena surfaces clear emotional and intent-based cues. This allows your team to respond with empathy, clarity, and the right level of urgency — all directly inside the ticket view.
💡 Why Sentiment Insights Matter
In IT support, the difference between a satisfied and frustrated customer often comes down to tone.
Helena bridges that gap by translating raw messages into clear emotional signals — so you know when to slow down, apologize, or reassure a user before jumping into technical details.
It’s not just about speed; it’s about improving customer trust and satisfaction (CSAT) with every interaction.
🧠 What Sentiment Insights Analyze
Every message in a ticket is evaluated across four key dimensions:
Insight Type | What It Means | Examples |
Emotion | The general emotional polarity of the message. | 🙂 Positive, 😐 Neutral and 😡 Negative |
Tone | The specific mood or attitude expressed in the message. | Satisfied, Neutral, Impatient, Confused, Frustrated, Angry |
Intensity | How strongly the customer feels the emotion (the emotional strength). | High, Medium, Low |
Root Cause | Why the emotion exists — what’s likely driving it. | Incomplete diagnosis, Communication gap, Delay, Unclear status, Repeated issue |
Helena displays these insights in the Sentiment Insights panel beside the ticket conversation.
📊 Sentiment Timeline
The Sentiment Timeline gives you a visual snapshot of how the customer’s sentiment has changed throughout the ticket conversation.
Each dot on the graph represents the detected sentiment (Positive, Neutral, or Negative) for a message or response.
🟢 Positive – Customer satisfied or appreciative
🟡 Neutral – Calm or purely informational
🔴 Negative – Frustrated, angry, or dissatisfied
As the conversation evolves, Helena maps these points chronologically, allowing you to track sentiment trends from first message to latest update.
How it works:
Each time you click the Helena icon, the system automatically re-analyzes all messages in the ticket.
The Sentiment Timeline graph refreshes in real time to reflect the latest conversation state.
This ensures the emotional data you see is always up-to-date and based on the most recent context.
🚀 How to Use Sentiment Insights
Step 1: Open a Ticket
Go to the Service Desk module.
Open the ticket you want to review.
Click the Helena icon on the right-hand chat panel.
Select the Sentiment Insights card from Helena’s options.
Step 2: Review Insights
Inside the Sentiment Insights tab, you’ll see a structured breakdown of how the customer’s messages are being interpreted:
Category | Description |
Emotion | Displays overall sentiment with emoji: 🙂 Positive, 😐 Neutral, or 😡 Negative. |
Tone | Shows detailed mood classification such as Satisfied, Neutral, Impatient, Confused, Frustrated, or Angry. |
Intensity | Indicates how strongly the emotion is expressed: High, Medium, or Low. |
Root Cause | Highlights why the emotion exists — for example, Incomplete diagnosis, Communication gap, or Delay. |
Helena updates these readings dynamically as new messages are added to the ticket.
Step 3: Apply Insights Before Replying
Use these insights to adjust your tone and message style before responding.
Detected Emotion & Tone | Recommended Response Approach |
🙂 Positive / Satisfied | Maintain a friendly, upbeat tone. Acknowledge the customer’s satisfaction and thank them for feedback. |
😐 Neutral / Confused | Keep communication clear and instructional. Provide concise updates or additional clarity. |
😡 Negative / Impatient, Frustrated, Angry | Lead with empathy and assurance. Apologize for any delay or issue, explain next steps, and confirm ownership. |
If intensity is High, it signals urgency — prioritize that ticket or escalate accordingly.
Step 4: Draft Smarter Replies with Helena
Once you’ve reviewed Sentiment Insights, switch to Suggested Reply.
Helena automatically adapts her draft tone to match the sentiment.
Example:
Detected: 😡 Negative Tone: Frustrated Intensity: High Root Cause: Delay
Helena Suggests:
“I completely understand this delay has been frustrating. I’ve just escalated this to our senior technician to ensure it’s resolved quickly. I’ll keep you updated as soon as we have progress.”
This keeps replies empathetic, relevant, and emotionally aligned with customer expectations.
🧭 Best Practices
Always check Sentiment Insights before replying — tone alignment reduces friction.
High-intensity negative emotions should be escalated quickly or responded to by senior techs.
Use the Root Cause field as a coaching cue for internal improvements (e.g., if many issues list “Communication gap”).
Pair this feature with Suggested Reply for automatically tone-matched AI responses.
Reopen the panel after new customer messages to see updated insights.
⚙️ Troubleshooting
Issue | Possible Cause | Solution |
No sentiment detected | No message received from customer user | Wait for a customer message or refresh ticket |