Release Summary - May 18, 2023
The following key features and improvements, along with bug fixes, have been released in Algonomy products in the release version 23.12 during May 06-May 18, 2023.
Personalization Platform
Introduced transparent background support for email recommendations
Email Recommendation feature introduces the ability to set a transparent background for the images rendered on your Shopper’s email client/browser. This setting can be activated by a new parameter you can set in the email open request called ‘omitBackground’=True. Using this new setting, you can set a background of your choice on your email template, and the product images and text will be rendered on that transparent background.
Jira: PLAT-3329
Enterprise Dashboard
Mapped page types to appropriate templates for Social Proof (New UX)
We have expanded the mapping of page types to appropriate templates for social proof (New UX). Now, you can create social proof for any of the pages as long as they fall into Item page (single product) or list page (multiple items). The following page types are now mapped to the appropriate templates: Add To Cart Page (Item Page), Brand Page (List Page), Cart Page (List Page), Category Page (List Page), Item Page (Item Page), and so on. Once you select the page type, social proof message types will show appropriately based on the template it is mapped.
Availability: Not yet available
Jira: ENG-25990
Add price question type to Guided Selling
Enhanced the Guided Selling experience for shoppers so that they can refine the results based on their budget. As a digital merchandiser, you can now create a question using price data in Guided Selling where the users can input a price range or a single value to filter the results.
Availability: Not yet available
Jira: ENG-25993
Find
Added skuField to productField mapping config for SKU affinity
We have enhanced our SKU Affinity feature by adding the ability to map skuField to productField in the complimentary JSON.
Jira: ENG-25947
Science
Moved Configurable Strategies ‘Top Sellers New Arrivals’ model to new model
The ‘Top Sellers New Arrivals’ configurable strategies model has been moved to the new model - Differential Top Sellers model.
Jira: ENG-25869
Recommend
Inconsistent exclusion of already purchased products in the last x day
The rule that was designed to filter out products from specified categories in the recommendation, if the end user has bought a product from one of those categories within the last 30 days was not working as expected. This rule was inconsistent in its behavior, occasionally failing to filter out relevant products. This has been fixed now.
Jira: ENG-25851
Data Engineering
‘Search’ performance visualizations in Find report
In this release, we have added performance visualizations for ‘Search’ in Find Performance Report for the following metrics, it’s a line graph on day-to-day basis.
-
Percentage visits using search.
-
Search queries with results Vs Search queries without results.
-
Search Vs non-Search metrics (CVR, AOV, RPV)
-
Average number of queries per visit.
-
Exit Rate (Bounce Rate)
Jira: ENG-25857
Tracking system visualizations for top 1000 users report
Introduced visualizations for the Tracking System (TS) Top 1000 Users report. This feature enables tracking and analysis of user behavior during instrumentation or client site changes. Visualize User Ids, RR User guide, and External Session Ids with the highest event counts.
Jira: ENG-25713
Other Feature Enhancements
The following feature enhancements and upgrades have been made in the release version 23.12 during May 06-May 18, 2023.
|
Jira # |
Module/Title |
Summary |
General Availability |
|---|---|---|---|
|
UPS > Platform: Enhance FTP upload scripts to process user Linking files with CSV/JSON format |
Enhanced the FTP upload scripts to process the user linking files with CSV/JSON format. To do this, the FTP upload script will pick the user linking files from the SFTP server and persist them to UPS Cassandra. |
18-May-2023 |
|
|
|
Platform>UPS: |
Currently, UPS feed only supports '^A' as the delimiter for user linking. As part of the initiative to integrate with AIQ (CDP), and accept bi-directional mapping from CDP to link Email user_ID with CRM user_ID (getting from the run time). UPS batch is now able to process user-linking files in CSV format with ',' as the delimiter between 2 IDs. Now it can accept a JSON user linking file also. |
18-May-2023 |
|
|
Data Engineering: Persist MVT rollup data and category dimension optimization |
Enhanced the MVT rollup process and optimized the Category dimension for improved data management. All 21 MVT rollup data points are now persisted to HDFS, enabling custom reporting. The Category, Category Placement, and Category PageType rollup data have been streamlined to include only the top 3 category levels in Cassandra, optimizing storage efficiency. |
18-May-2023 |
|
Recommend: Size based personalization issue – need to change "size" attribute name |
Using "size" as the SKU attribute for the size-based personalization was causing issues for some customers. This happened as the customers were using a global attribute named "size" for facets. Hence changed SKU name to more unique name "perso_size_sku". |
18-May-2023 |
|
|
Find: SearchSchemaException in Search Service |
Noticed production issue in SearchSchemaException in search service as resolveField was producing exception if field was not matched. The issue has been resolved now |
26-May-2023 |
|
|
Find: NPE while loading VersionedLanguageSchema |
When we ran some tests in QA search-service using the following request we found an NPE: http://10.3.252.38/searchservice/query/1762/content?q=Crestron&lang=lv&eqDefaultWeight=1&debug=solrDebugAll In QA Solr there's no 1762-content-lv-* collection but we have 1762-catalog-lv-* and the searchservice is able to reach this point. The NPE is happening at line 86 of the CoordinatorLanguageSchemaProvider The issue has been resolved now. |
26-May-2023 |
|
|
Streaming Catalog: SiteConfiguration cache produces 'none' when http site configuration is unreachable |
SiteConfiguration cache was producing 'none' when http site configuration was not reachable. Catalog NCA uses the SiteConfigurationsCache, which skips when None is found. Add the following consul config in all Streaming Components: site-configurations-cache.refresh-interval = 5m And deployed steaming components 1.13.4. The issue has been fixed now. |
26-May-2023 |
|
|
UPS: UPS Compactor is taking more than 48 hrs. to complete |
UPS compactor was taking more than 48 hours to complete the process, and a few instances were stuck in the compaction process and were not able to stop. As part of the fix, we have Resized HashMap and Optimized view events. The issue has been fixed now. |
26-May-2023 |
Bug and Support Fixes
The following issues have been fixed in the release version 23.12 during May 06-May 18, 2023.
|
Jira# |
Title |
Summary |
General Availability |
|---|---|---|---|
|
Science: Configurable Strategies - View results should be de-duplicating by short name |
A client was not getting de-duplicating products on the View Results page in Configurable Strategies. However, it is expected that if the “short-name de-duplication” option is enabled, then the View Results should be de-duplicating products, unless the option "exclude system and user rules" is selected. This issue has been fixed now. |
18-May-2023 |
|
|
Data Engineering: Graph not refreshing for Total Raw call count metric in Find API Report |
The graph was not refreshing when the user selected the Total Raw Call Count metric in Find API Report. This issue has been fixed now. |
18-May-2023 |
|
|
Find: Unable to create any solr collection for streaming clients in QA |
When checked in language service it was found that the plugin file was incorrect. Correct file has been uploaded and the issue has been fixed. |
18-May-2023 |
|
|
Find: Find-py-scripts are not launching release script on entry point |
Find-py-scripts were not launching the release script on the entry point while executing the script. It was docker issue, the docker issue has been resolved and the docker find-py-scripts image is able to run the release script and unload itself when it completes. |
18-May-2023 |