Skip to main content

Release Summary - Apr 20, 2023

The following key features and improvements, along with bug fixes, have been released in the Algonomy products in the release version 23.10 during April 07- April 20, 2023.

Enterprise Dashboard

Display ‘Do not Recommend’ rule categories only for rules that are blacklisted

The display of "recently purchased" or "pending purchase" options in the 'Do not Recommend’ rule has been fixed to only apply to blacklisted rules. Previously it was displayed for all rules.

The "What" column on the list page now shows entries for the 'Do not Recommend rule' saved with "Filter products from categories the user has already purchased".

Jira: ENG-25781

UI change in Advanced Merchandising for editing rules

The new Advanced Merchandising UI has been redesigned with a few changes, primarily aimed at eliminating the need for scrolling down the page to view all recommendations for each seed product. With the updated UI, users can quickly flip through the seed products and see all recommendations,  enabling them to evaluate the rule more effectively.

Jira: ENG-24310

Creating multiple User Affinity Configurations

Users can now create multiple user affinity configurations and select the appropriate configuration to use in Configurable Strategies for each strategy. This enables them to apply diverse strategies with different user affinity configurations to play on various page types, channels, or for other context or user segment options.

Jira: ENG-22568

Other Feature Enhancements

The following feature enhancements and upgrades have been made during April 07- April 20, 2023.

Jira #



General Availability


Data Engineering, Enterprise Dashboard:

Enabled Find API Report in the Reports Menu

Find API Report has been enabled in the Report menu, providing key metrics to record the FIND API call counts.




Integrated Omni channel API with common status framework

Omni channel Order API has been integrated with the common status framework for async reporting of the request status to report:

An overview of a request, including number of orders, success, and failures.

The reason for the failure and order IDs of the failed orders.




Creating a timing log for each component in the Search service

With approximately 30 query processors, the Search service has enhanced the processing speed of every Find request by adding timing information for each individual component.



Data Engineering & Engage:

Dynamic field support for new variables in Dynamic Experience Templates

When a new template is created for a variation, the layout JSON is sent as the payload of an API call to the backend. However, only 5 fields are accepted while any additional fields are disregarded. Consequently, the fields are hard coded.

Dynamic Experience templates now support dynamic field creation for new variables.



Data Engineering:

Changing the metrics ‘Rec Views’ and ‘Rec Clicks’ to ‘Views’ and ‘Clicks’ respectively in the Real Time Report

Real Time Dashboard metrics have been changed to ‘Views’ and ‘Clicks’ instead of ‘Rec Views’ and ‘Rec Clicks’ to better assist merchandisers in real-time reporting.




Adding option to Merchandising Rules to remove products already viewed by the user

The user interface has been improved to make it easier for Merchandisers to use the popular products model and filter out items that the user has already viewed. Additionally, there is now an option to experiment with new products in the placement.




Creating Default Configuration for multiple User Affinity Configuration

When Merchandisers create multiple user affinity configurations and tag them as default, the system may not know if the configurable strategy was linked to the correct user affinity configuration. This is because the configuration was tagged as default and it's unclear whether it best matches the configurable strategy.

To address this issue, Merchandisers can create a 'Default' user affinity configuration. This allows them to edit the value of the default at any time and link it to the configurable strategies that are marked for the default. The name 'Default' cannot be changed or deleted, and a backend API is required to create a configuration with this name that cannot be deleted.




Changing Find API response for Status: Warning message even though user is not set

A client encountered an issue while rendering search results on the page based on the value of 'status' returned in the response. If the value of 'status' was 'OK', the products were added to the page. However, if the user was not logged in, Find returned 'status=Warning', which caused an issue for the client.

This issue has now been resolved, and the client can see 'status=OK' in the Find response even when the user is not logged in. Additionally, the search service parameter-related information has been moved from 'Error' to 'Search service debug' to provide more comprehensive information to the client.



Data Engineering, Enterprise Dashboard:

Find API Count Reporting


Customers can now view the Find API count reporting as part of the portal to track the number of API calls made on a daily basis. The Find API report provides insights on the daily query count for different API methods and enables clients to analyze daily trends.

The following metrics are available in the Find API report:

Total Call Count

Total Raw Call Count




Delete support in BuildFTP for Segments, User attributes and User links Feed types


Support has been removed from BuildFTP to enable error-free processing.

The following changes have been made:

BuildFTP will now process user records and remove all content for the specified feed type upon receiving an empty payload for the userID.

This change is applicable to all feed types, including Segments, Columnar Segments, User Attributes, User Links, and both Full & Delta Grouped Links.




Performance improvement on CategoryBasedReplenishmentModel

The 'CategoryBasedReplenishmentModel' request was taking longer than expected due to the selection of a full product catalog, which includes all products and attributes. Since each product only looks at category membership, this caused unnecessary delays. With the latest update, the request has been modified to build local catalog data that excludes attribute data and runs a single query that only returns category data. This change significantly improves performance.



Advanced Merchandising, Discover, Find, and Recommend:

Performance improvement on Legacy CatlaogReader

The ‘CatalogReader’ used by Catalog model builds, Discover models, Find models, and Advanced Merchandising has become outdated and inefficient. The latest version significantly improves performance by collecting all the data at once in a product scan, rather than collecting the data from the client-side. This change reduces request completion time.




Statistics events data recovery through Avro log data

System map for Avro data is now being read in and converted to statistics service events. 25-Apr-2023

Bug and Support Fixes

The following issues have been fixed during April 07-April 20, 2023.




General Availability


Dashboards, Identity & Access Management:

Model Browser access if the URL is known


Previously, all users could directly access the 'Model Browser' if they knew the URL. However, the 'Model Browser' option should only be accessible to users who have been assigned the 'RR Super User' role or any of the internal roles. This issue has been fixed now.



Dashboards, Identity & Access Management:

Unable to see the status for Offline feeds

A client was unable to view the status of Offline feeds. This issue has been fixed now.




Issue with Brand Filter and Attribute Filter in Recommend

A client faced an issue while creating a Strategy rule and Restriction Rule, where the Brand Filter and Attribute Filter were not displaying the expected results. This issue has been fixed now.




External Facets Filtering Issue


A client encountered an issue with external facets filtering when applying the size filter using the FIND method and sending the products via the POST method. The sizes of other products were being shown in the response as well in the external facets. This issue has been fixed now.



  • Was this article helpful?