Accelerate time to key insights
Bridging quantitative and qualitative insights with ML
UserTesting’s AI and ML implementation is designed to remove repetitive tasks while allowing humans to focus on nuanced, high-cognitive tasks. Our proprietary ML models are trained on massive volumes of unique conversational and behavioral data that have been processed through the UserTesting platform over the past decade.
From participant quality control to sentiment and intent analysis, we embed our ML models throughout the insights discovery and analysis workflows in order to accelerate time to critical insights.
Interactive Path Flows
Built on recent research in data mining, the Interactive Path Flow aggregates interaction data across multiple participant sessions to visualize the customer journey, surface unexpected behaviors, and navigate you to key moments in the customer journey.
Sentiment and Intent Path
ML-generated Sentiment Path automatically display positive and negative sentiments corresponding to specific points in the customer experience.
Intent Path identifies over 100 intents and behaviors like browse, search, or comparison to discover which actions correspond to an intended goal like add to cart.
Hover over the path to quickly jump to the relevant video clip for more context.
Instant insight drives efficient post-test analysis by automatically identifying patterns, anomalies, and correlations across contributor sessions. Findings can be mapped to customer intent or goals in the Interactive Path Flows using custom tags.
Similar keywords are automatically grouped based on overall sentiment (positive, negative, or neutral) to identify themes. Highlight reels associated with each keyword are grouped together so that you can directly drill in and understand the why behind each sentiment.
Quickly gain visibility into customer interaction patterns with Click maps. Discover which web page elements they’re engaging with or get a detailed list of every interaction they have with an individual screen element. Click maps will aggregate all contributor data into a single visualization for faster analysis and sharing.
Audience QA and performance management
Each contributor—a participant in the UserTesting network—is evaluated before they join the UserTesting Contributor Network to ensure high-quality session results. Machine learning models are also employed to accelerate the acceptance and inclusion of high-quality contributors that regularly receive 4 or 5 star reviews from customers like you.