Make Smarter, Data-driven Decisions
with AI-based virtual assistant and conversational analytics.
The Kore.ai Experience Optimization Platform provides real-time, actionable insights with all the data and visualization tools you need to measure and analyze your virtual assistant performance from a centralized dashboard.
The analytics allow you to build a deeper understanding of your customers and your employees and enable you to continually measure, analyze, and refine your virtual assistant strategy.
Virtual Assistant Engagement
Better understand how users interact with your assistants, allowing you to increase user engagement, acquisition, monetization, and retention through actionable analytics.
The platform provides an abundance of data that can be mined to improve performance, including:
- Tasks performed
- Alerts sent
- Total chats and sessions
- Active users
- Demographics
- Top tasks
- Top assistants within an account
- Session benchmarking
Get as granular as you want with the ability to filter information based on categories like duration, language, and channel.
Conversational Analytics
Build better, user-focused conversation flows that create engaging, memorable experiences for your core audience.
Conversational Flows
Gain a deeper understanding of user behavior. Conversational flows map popular user paths, tasks, and exit points within a visual context, helping to surface patterns, trends, and correlations that might go unnoticed using text-based data analysis methods.
The user Utterances are clustered based on similarity, so you can quickly review if the correct utterances are going to the right intents.
Virtual Assistant Transcripts
Track the entire user lifecycle. The platform allows you to view the whole chat history of user sessions, number of interactions and monitor current conversations. It protects user privacy by limiting access to only those with the appropriate privileges and automatically redacting personally identifiable information (PII).
Sentiment Analysis
Gain deeper insight into users’ interactions – whether they are positive or negative – and provide better service to employees and customers.
The platform programmatically predicts user emotions by extracting keywords and topics from user utterances and evaluating them against common user sentiments such as joy, fear, and anger. It generates a score ranging from -3 to 3 based on the intensity of the emotions detected.
Functional Analytics
Get the data you need to train and improve your virtual assistant performance, including how well they are identifying and executing tasks. Well-trained virtual assistants mean more conversations, efficient task execution and better productivity.
Intent Matching
Give your bots the skills they need to meet user expectations. Our platform records all utterances that were successfully matched to a unique intent and all utterances that did not match to unique intent.
All utterances are tagged with detailed meta information, such as channel, intent detection logic, language, user, and more. This information can be filtered by intent, user, date, and language.
Utterance Grouping
Spend less time combing through chat scripts to find the data you need to effectively train your virtual assistants. Our platform groups all utterances based on ML models and semantic similarity, including stop words, spelling corrections, and nouns.
Entity values are also extracted and replaced with placeholders. This allows you to quickly identify patterns, select large sets of similar utterances at once, and train your assistant.
Task Failures
Effectively identify and troubleshoot assistant task issues. The platform records and categorizes all instances where a virtual assistant fails to complete a task, as defined by the developer during the dialog creation process.
All contextual information is retained, along with the dialog flow; you can review the chat transcript or service response JSONs at any time to better understand the underlying causes. These include tasks aborted, alternate tasks, agent handoffs and authorization failed.
Advanced Analytics and Visualization
Segmentation
Get in-depth information about your virtual assistants with meta-tagging and segmentation capabilities. Tag the conversations or sessions with business-specific keywords and call platform functions at any time in a dialog to append, delete, or retrieve a tag.
Post tagging, conversational and functional analytics can be filtered based on your business rules, allowing you to create effective business reports that match your requirements. Sample tags include customer complaints, flagging, source and more.
Custom Reporting
Create custom reports and custom data visualizations to fit any enterprise or development reporting needs. All standard dashboards and data visualizations can be customized according to your unique use cases and sliced and diced to discover more insights.
Data from the reports can be presented as graphical widgets on a dashboard and exported into external formats, especially useful for business reporting and ROI calculators.
Data Extraction
Quickly and easily extract and embed conversational and functional analytics data into your analytics software. Our platform exposes this data via custom tags.
