This release is all about deeper conversation context, secure handling of personal data, and a better understanding of how customers interact with your chatbot. Three features that make your team’s workflow more structured and your analytics more accurate.
Conversation Attributes: capture context directly in the chat — inquiry type, order number, category, quality rating
Personal Data Anonymization: flexibly hide names, phone numbers, emails, and contact attributes based on user roles
Chatbot Scenario Logging: track events within chatbot flows and visualize customer behavior in Metabase
Conversation Attributes
Conversation attributes are structured fields linked to a specific interaction, allowing you to capture any important context directly in the chat.
Every customer inquiry contains more than just text. Now you can store key details прямо in the conversation: quality rating, inquiry type, product or service category, order number, and anything else relevant to your business.
Custom conversation attributes are tailored to your workflows and displayed where your team needs them. This means all context is always at hand — no switching between systems and no loss of important details. Agents resolve inquiries faster because they don’t need to ask повторні questions — all necessary data is already captured. Analytics becomes more meaningful: reports on categories, reasons for обращения, and service quality reflect the real picture, not just the number of conversations.
How to set up: go to Settings → Custom Attributes → Conversation
Personal Data Anonymization
Personal data anonymization allows you to hide sensitive contact and conversation fields from employees who don’t need access to them.
Not every team member should see full customer details. This feature enables flexible control over what data is visible to each role in the system.
You can hide:
- Customer name
- Phone number and email
- Description and notes
- Contact tags
- Custom attributes (at both contact and conversation levels)
Each user sees exactly what they need to do their job.
This is especially important for businesses handling sensitive personal data or operating with multiple roles and access levels.
How to set up:
- Go to Settings → Roles & Permissions
- Create a new role or edit an existing one
- Enable anonymization for the required fields
Chatbot Scenario Logging
Chatbot scenario logging records events that occur during a customer’s interaction with the bot — from the first message to completion or handoff to an agent.
Typical tracked events include:
- Language selection at the start of the conversation
- Menu navigation (clicks, submenus, back actions)
- Text input responses
- Self-service flow start and result (successful or failed)
- External service/API calls
- Handoff to an agent with reason tracking
- Conversation завершення with reason (manual close, timeout, off-hours, CSAT flow)
If you use a chatbot, this update helps you truly understand how customers interact with it — which language they choose, which menu paths they follow, where they drop off, and why they escalate to an agent.
All collected data can be visualized in Metabase, enabling you to build dashboards and track customer behavior over time. This helps improve chatbot scenarios, reduce agent workload, and increase self-service efficiency.
Logging is configured individually. Detailed documentation with parameters and examples is available in our knowledge base. To enable this feature, please contact your account manager.
Who is this for?
- Support teams with high volumes: structured workflows and precise analytics
- Businesses handling sensitive data: flexible access control and compliance
- Companies using chatbots: insights into drop-offs and optimization opportunities
- Teams with complex role structures: different access levels without extra integrations
FAQ
What are conversation attributes?
Structured fields linked to an interaction (e.g., inquiry type, order number, category, quality rating). They help capture context and improve analytics.
How are attributes different from tags?
Tags are free-form labels. Attributes are structured (text, number, date, list), making them more accurate for reporting.
What data can be anonymized?
Name, phone, email, notes, tags, and custom attributes at contact and conversation levels — depending on role settings.
Does anonymization affect agent performance?
No. Agents still see all necessary information. Irrelevant data is hidden, reducing noise.
What chatbot events can be logged?
Language choice, menu navigation, drop-off points, handoffs, and conversation completion reasons.
Key chatbot metrics to track?
Self-service resolution rate, drop-off points, popular flows, and escalation frequency — all visualized in Metabase.
How to enable the features?
Conversation attributes and anonymization can be configured in Settings. Chatbot logging is set up individually via your account manager.
