Overview
The 25.2 release brings significant enhancements to the Journey Discovery Analytics experience and graphs functionality as well, designed to help users extract deeper insights, streamline analysis, and make more informed decisions. These updates reduce the manual effort required to understand customer journeys and graph behavior by introducing more intuitive interfaces, proactive guidance, and visualization. Whether identifying performance bottlenecks, exploring customer paths in greater depth, or interpreting complex graph logic, the new capabilities make it easier for both technical and business users to uncover meaningful patterns and take timely action.
Journey Insights: The new Journey Insight cards in Journey Discovery Analytics provide users with proactive, data-driven summaries of how their journeys are performing. When landing on the Analytics page, users now see a set of Insight Cards highlighting key trends, anomalies, and areas for optimization, like high drop-off rates, unusually long wait times, and repeated user actions.
Each card not only surfaces meaningful metrics but also suggests potential actions to take, enabling faster decision-making without manual analysis.
Users can toggle between focused, detailed, or contextual views to explore the most significant patterns, such as steps with the longest wait times, highest repeat rates, or journeys with the lowest completion. Each view also includes supporting context like total volumes, number of profiles, and total interactions to help assess overall impact.
Insight Metrics and How They Help:Avg. Time on Step Change: Compares the average time spent on steps in the last 30 days vs. the last 6 months average. This helps identify recent shifts in user behavior.
Avg. Time on Step: Highlights the step with the highest average time spent by users over the past 30 days. This may indicate friction or user confusion.
Step Stall Rate: Reveals hesitation points where users pause for longer periods, but also indicates fallout in the journey, prompting teams to review content or system behavior.
Step Repeat Rate: Detects where users loop back or retry steps, indicating steps need clearer instructions or streamlined logic.
Journey Completion Rate: Displays the journey with the lowest completion rate in the last 30 days, helping identify where users are dropping off. This is the only insight based on the journey; others are on individual steps.
Journey Step Focus View: Journey Step Focus View allows users to anchor the Sankey visualization to a specific journey step. Then it shows what typically happens immediately before and after that step across all occurrences in the selected date range. Unlike traditional Sankey views, which follow a fixed path sequence (Step 1 > Step 2 > Step 3, etc.), Focus View ignores the step position and instead aggregates all instances of a selected step, showing the most common steps taken immediately before or after it regardless of where it appears in the journey.
By default, the view shows the most frequently occurring steps immediately before and after the anchored step, with the option to “View More” for a full list. This allows users to understand how often each surrounding action occurs about the anchored step, supporting use cases like troubleshooting drop-offs or optimizing engagement flows. The Focus View retains filters, outcome/stall toggles, and step-level tooltips. Users can easily switch between Focus View and the standard Sankey View from the toolbar.
AI Graph Explainer: The AI Graph Explainer offers a conversational, plain-language interface to help users understand complex graphs, uncover insights, and debug more efficiently. Admins can enable the assistant at the org level, securely manage API keys, and test configurations. The assistant functions seamlessly across graphs, resetting automatically when switching graphs, handling invalid graphs gracefully, and excluding managed graphs unless accessed by users.
These capabilities make it a practical tool for configurers looking for faster, more intuitive graph analysis. Graph Explainer is powered by Xponent Intelligence and must be activated for your organization.
Please contact your admin to enable the Graph Explainer feature.
Bug Fixes
We have fixed the following bug, ensuring these functionalities now work as expected:
KIT-14981 - Improved message consumption speed in the Profile Ingestion Service (PIS) when processing large data volumes from DMTS. Previously, high-volume imports. Now, optimizations in message handling ensure faster throughput, reducing delays during large-scale profile ingestion.
KIT-15014 - Fixed an issue where the pagination component displayed incorrect record ranges and empty results on the last page. For Example: With 20 total records and a page size of 10, the last page incorrectly showed “21-20 of 20” and no records. Now, the last page correctly displays the expected range (“10-20 of 20”) and all records load as intended.
KIT-15024 - Project version creation did not use a transaction, leading to failed rollbacks and blocking future version creation. Now, a Sequelize transaction is correctly applied to ensure proper rollback and cleanup during errors.
KIT-15158 - The “Analytics” and “Profile View” links in the profile detail view now appear at the correct size with the missing chevron icon restored. Previously, the text was too small and the navigation icon was missing, making the UI inconsistent.
KIT-15280 - The Kafka adaptor now fails gracefully and logs a clear error when an invalid Kafka topic is configured. Previously, it appeared to run successfully despite connection issues, making it hard to detect misconfigurations.
KIT-15318 - Fixed an issue where the file upload button was hidden when the browser zoom level exceeded 125%, due to a persistent grey bar at the bottom of the screen. Now, the interface allows scrolling so the upload button remains visible and accessible regardless of zoom level.
KIT-15362 - Metadata fields can now be assigned as inputs to adaptor nodes. Previously, attempts to assign metadata to adaptors were ignored or failed silently, while the same inputs worked on base nodes.
KIT-15440 - Resolved an issue where a batch listener deployment would incorrectly move to a critical state if all files were removed during deployment. Now, deployments remain stable even when no source files are available.
KIT-15501 - Fixed intermittent profile import failures during large CSV uploads in the CSA flow. Now, imports handle temporary backend disruptions more gracefully, ensuring all valid records are processed without missed updates.
KIT-15526 - Transaction log searches now return relevant information instead of unrelated connection errors. Previously, filtering by transaction ID sometimes showed backend connection issues, making it hard to trace actual transaction activity.
Supported Versions Update
As part of ongoing efforts to maintain a secure and modern platform, we’ve completed several backend upgrades. These include:
Runtime Environment Updates: Core components have been upgraded to the latest language versions to ensure improved performance and long-term support.
Cloud SDK Alignment: AWS SDKs have been standardized across key systems to streamline integration and support future scalability.
These changes are foundational and do not impact customer-facing functionality or require any action.