Release Summary 25.02 | Jan 23, 2025
The following key features and improvements, along with bug fixes, have been released in Algonomy CXP products in the release version 25.02.
Recommend
Enhanced Wishlist Model in Configurable Strategies
The User Wishlist History model is now available in Configurable Strategies, enabling optimization managers to recommend products saved in shoppers' wishlists for effective re-engagement. This model retrieves items based on the most recent wishlist view, accounting for additions and removals, and supports multiple wishlists.
Key features include configurable seed options (e.g., Brand/Category Context, Brand/Category Affinity), sorting by User Affinity or Smart Shuffle, and filtering options such as Diversify by Category or Brand. The Preview page has been updated to display the model name, seed, sort, and filter details, with the user ID field always visible. Additionally, items are naturally sorted by recency, ensuring relevance in recommendations.
Enterprise Dashboard
UI Updates for Advanced Merchandising Strategy Mode
The Advanced Merchandising UI has been enhanced to support the recently introduced "use AM as strategy" mode. When this mode is enabled, certain rule options are predefined to ensure consistency. Specifically, Enable system filters and rules and Remove Duplicates are selected by default, while Backfill remaining slots with recommendations is deselected.
These predefined options are now disabled in the UI to prevent changes, ensuring Advanced Merchandising functions optimally as a strategy.
Jira: ENG-29631
Enhancements to User Affinity Configuration Charts
We have improved the charts in User Affinity Configuration to provide clearer and more concise visualizations. Elements with a score of 0 are now hidden to declutter the charts and emphasize relevant data, especially for categories using the new tier options.
Additionally, chart values are sorted by their combined affinity score, with bars displayed in descending order from left to right. For custom attributes, only the attribute values (e.g., "M" for size) are shown on the labels along the x-axis, simplifying the visualization and making it easier to interpret shopper affinities at a glance.
Jira: ENG-29473
Ensemble AI
Configure Complementary Colors at Site Level for Ensemble AI
Merchandisers can now view, update, and configure complementary colors used in the Ensemble AI model, enabling precise control over color combinations for scoring and ranking ensembles. This feature allows users to define complementary color configurations specific to their site.
Key functionalities include:
-
A Complementary Color Configuration table that displays:
-
Primary Color
-
Count (Number of complementary colors available)
-
Complementary Colors (Editable options)
-
-
Edit Options:
-
Modify existing complementary colors.
-
Add new complementary colors from a predefined list.
-
Remove unwanted complementary colors.
-
Save changes seamlessly.
-
-
Access the configuration table through the "Complementary Color Configuration" option in the Ensemble AI style definition, presented as a popup window.
Jira: ENG-29013
Dynamic Template for Ensemble AI Experiences
We have introduced a new dynamic experience template for Ensemble AI to streamline the creation of experiences for merchandisers. The "Ensemble AI" template is now available as part of the existing template library, eliminating the need to copy code from other experiences.
The template includes two predefined variables such as Element ID and Similar Products Strategy Name.
Jira: ENG-29037
Social Proof
Export Dynamic Experience Report Tables
Merchandisers can now download Social Proof and Dynamic Experience report tables directly from the portal, enabling easier reporting and data sharing. The export functionality supports de-aggregated versions of the reports, allowing users to download detailed metrics, including lift and confidence, in CSV or Excel formats.
Jira: ENG-29486
Other Feature Enhancements
The following feature enhancements and upgrades have been made in the release version 25.02.
|
Jira # |
Module/Title |
Summary |
General Availability |
|---|---|---|---|
|
Data Engineering: Enhanced Consistency in Findability Metric Calculations
|
We have addressed discrepancies in the calculation of the findability metric between the Find Report and the Find Search Terms Report. The differences arose due to varying click calculation methods: Find Search Terms Report: Considers only one click per viewguid for calculation. Find Report: Includes all clicks associated with a single viewguid in the calculation. These differences affected the reported clicks and clicks on the first page, leading to mismatched findability metrics. The calculations now work as intended, ensuring transparency and accuracy in reporting. |
09-Jan-25 |
|
|
Enterprise Dashboard: Updates to Advanced Merchandising Configuration |
The "use am as strategy" Site Configuration has been moved to the Advanced Merchandising section for better organization. Additionally, the following label changes have been made for clarity: "use Advanced Merchandising" is now labeled as "enable Advanced Merchandising." "use am as strategy" is now labeled as "use AM as a strategy." |
23-Jan-25 |
|
|
|
Enterprise Dashboard: Static Files for Guided Selling in Engage Repo |
The Guided Selling page has been updated to use static files instead of changescripts for loading the default template. Default templates now load directly from static HTML, CSS, JS, and JSON files, ensuring improved reliability and maintainability.
Custom templates saved with experiences or variations remain unaffected, and the existing changescripts have been removed. |
23-Jan-25 |
|
Find: Accurate Error Logging for Solr Timeout Errors |
Solr timeout errors in Solr 9 are now correctly categorized under the SEARCHSERVICE_SOLR_TIMED_OUT_REQUEST error bucket instead of being misclassified as EB_UNKNOWN. |
23-Jan-25 |
|
|
Find: LLM Async Implementation for Improved Performance |
The LLM Bulk API for Related Searches has been updated to support asynchronous operations, significantly enhancing overall performance.
|
23-Jan-25 |
|
|
Streaming-recommend: Region Support in Streaming Recommend |
Streaming Recommend now supports region-specific overrides for attributes such as in_stock, start_date, sale_price, and price, allowing product recommendations to reflect accurate regional availability and pricing. These attributes can be dynamically configured and verified in the streaming item store. Using the region parameter (rid) in recommend requests, the system ensures that product data is tailored to specific regions, improving the relevance and precision of recommendations. |
23-Jan-25 |
|
|
Updated MongoDB Date Query Logic for QueryWriter Jobs |
The QueryWriter and related jobs have been updated to address an issue where search terms from the same day as the job execution were being excluded. This was caused by the use of a < operator for the upper date bound in MongoDB queries, which inadvertently omitted documents with timestamps at the exact upper bound. To resolve this, the date query logic has been updated to use a <= operator for the upper bound, ensuring all documents within the specified range are included. This change eliminates edge cases and guarantees accurate data retrieval in QueryWriter jobs. |
23-Jan-25 |
|
|
Social Proof: Enhanced Social Proof Reporting |
Social Proof reporting now provides detailed metrics, distinguishing between total visits for each variation and visits where a Social Proof message was shown. This enhancement allows Digital Optimization Managers to evaluate the effectiveness of Social Proof messages and their impact on user behavior. The reporting includes rollup metrics for visits with Social Proof messages and corresponding performance data, alongside metrics for visits without messages. The Dynamic Experience Reporting API has been updated to deliver this granular insight, supporting all Social Proof variations. Note: The associated UI task for this feature is still pending. |
23-Jan-25 |
|
|
Find: Filtering Bot API Calls in Find API Call Count |
The Find API call count now excludes bot-generated API calls (flagged as isRobot=true) from billing calculations, ensuring clients are billed only for valid user-generated requests. Bot calls are identified and counted separately, providing improved transparency in API call reporting. This enhancement applies to both individual API calls and those calculated over 3-second intervals, delivering accurate and fair reporting for Digital Optimization Managers. Note: The associated UI task for this feature is still pending. |
23-Jan-25 |
Bug and Support Fixes
The following issues have been fixed in the release version 25.02.
|
Jira # |
Module/Title |
Summary |
General Availability |
|---|---|---|---|
|
Guided Selling Progress Bar Calculation and Animation Issues |
We have resolved issues with the progress bar and screen animations in the Guided Selling flow on the REI site for the Activity-Based Quiz. The progress bar now dynamically adjusts based on the user's selected quiz path in branch mode, accurately reflecting progress only for screens displayed in the current path. Additionally, backward movement of the progress bar has been eliminated. Screen animations now move consistently forward when progressing to the next question and backward only when the back button is used. |
23-Jan-25 |
|
|
Data Engineering/ Enterprise Dashboard: External Product ID Display on Real-Time Report |
We have updated the Real-Time report to display the external product ID instead of the internal product ID, providing more meaningful information for merchandisers. This change applies to both the Dashboard home page (Real-Time report section) and the Products tab within the Real-Time report. |
23-Jan-25 |
|
|
|
Enterprise Dashboard: Content Campaign Report Deaggregation Issue |
We have resolved an issue in the Content Campaign Report where the aggregate option would reset when attempting to deaggregate data by campaign. Users can now successfully select campaigns and deaggregate data without the option being reset.
|
23-Jan-25 |
|
Enterprise Dashboard: Breadcrumb Categories Displayed Out of Order on Product Catalog Page |
We have resolved an issue where category breadcrumbs on the Product Catalog page were displayed out of order, disrupting the parent-child hierarchy. The breadcrumbs now correctly reflect the hierarchical order, starting from the top-level parent category down to the selected category. |
23-Jan-25 |
|
|
Enterprise Dashboard: Unable to Delete Attributes from User Affinity Configuration |
We have resolved an issue where attributes could not be deleted from the User Affinity Configuration. Users can now successfully remove attributes, save the configuration, and see the updates reflected as expected. |
23-Jan-25 |
|
|
Enterprise Dashboard: Guided Selling: Bugs related to multi language |
Resolved several multi-language bugs in Guided Selling: Improved translation template handling. Enhanced special character support for Mac users. Fixed issues with saving and displaying modifications. |
23-Jan-25 |
|
|
Engage: Engage Affinity Sorting for Enhanced Tags
|
We have resolved issues with Engage affinity sorting for contents associated with Enhanced Tags and parent-level categories. Contents linked to parent-level categories (e.g., Tier 2 categories) are now correctly returned in affinity-based campaigns, ensuring all relevant content is included. Additionally, when multiple contents have the same affinity score, the system now randomly selects from those options, preventing repetitive results and delivering varied content placements. |
23-Jan-25 |
|
|
Variable Name for Location Selector in Dynamic Experience Campaigns |
We have resolved an issue with the variable name used for location selection in Dynamic Experience campaigns. The variable has been updated from "element_selector" to "message_location_selector" to ensure accurate campaign rendering based on priority and location. |
23-Jan-25 |
|
|
Graph Visualization Error in Dynamic Experience Report |
We have resolved an issue where the graph visualization in the Dynamic Experience Report failed to load correctly, displaying the error: "All series on a given axis must be of the same data type." The graph now renders properly with the correct x-axis and y-axis format, ensuring users can view the data as expected. |
23-Jan-25 |
|
|
Email HTML Generated in Test Drive is Not Correct |
We have resolved an issue where the email HTML generated in Recs Test Drive incorrectly pointed both anchor and image tags to our image server. The anchor tag now correctly points to the click URL. |
23-Jan-25 |
|
|
Find, Streaming Catalog: Engine sent null values of enrichment woc fields |
We have resolved an issue where the engine was sending null values for enrichment WOC fields to engine.out, even though the enrichment view contained the correct data. Enrichment is now functioning as expected after resetting the offset. |
23-Jan-25 |
|
|
Streaming Catalog: Enrichment Sidekick fails to delete dataset and crashes |
We have resolved an issue where the enrichment sidekick service would fail to delete datasets and crash, leading to service instability. |
23-Jan-25 |