Release Summary - Sep 07, 2023
The following key features and improvements, along with bug fixes, have been released in Algonomy products in the release version 23.20 during Aug 25 - Sep 07, 2023..
Enterprise Dashboard
Enhanced Localization: Returning Language-Specific Values with Language Parameter
In this release, we've introduced an enhancement that enables the system to return localized values when a language parameter (lang) is set. This means that product names, descriptions, URLs, category names, and other attributes using language as the localization type will now be displayed to end-users based on their selected language. This feature applies to various APIs, including recsForPlacements, recsForPlacementsContext, p13n, and recs in email.
This update ensures a consistent and seamless user experience, allowing you to cater to a diverse audience by delivering content in their preferred language.
Jira: ENG-25805
Streamlined Client Whitelisting
We've improved our client readiness process by streamlining domain whitelisting through Akamai. Previously, each client site required manual whitelisting, potentially causing delays. In this release, we've automated the process for our QA and Production environments, reducing wait times. We've also explored alternative whitelisting methods for client sites, enhancing flexibility and efficiency in onboarding.
Jira: ENG-25992
Exclude Specific Sites from Visit Drop Alerts
In this release, we've enhanced our alerting system to provide more precise and relevant notifications. To ensure you receive only pertinent alerts, we've introduced an exclusion criteria for sites. This means you can exclude certain sites, like those in sales or listen mode, from the alerting process, ensuring your alerts remain focused on the most critical areas.
We've also improved the alerting logic. Now, we compare the average visits over the last two weeks with the average visits on the same day over the last five weeks. An alert is triggered only when both criteria show a 50% drop, providing more accurate and actionable alerts. Furthermore, we've made alerting configurations more flexible, allowing you to customize alerts for drops in both visits and attributable sales.
Jira: ENG-26304
Template for Applying Social Proof to Recommendations
We have introduced an enhancement that empowers eCommerce product managers to further engage users and boost click-through rates through the inclusion of social proof messaging within product recommendations.
This feature enables you to show more than one social proof message on the same page type, ensuring that users see messages relevant to their current journey. The importance of these messages is determined by placement type and template, allowing you to deliver a customized experience.
Configuring Social Proof Messaging is simple, with templates aligned with your existing list page configurations. You can specify where messages are applied by entering placement names, ensuring precise control over message placement. This enhancement empowers you to provide users with context-aware content, improving their shopping experience and increasing engagement.
Jira: ENG-23288
Recommend
Improved Rule Editing
In this release, we've enhanced the rule editing experience by making it more intuitive. For existing rules created before version 23.18, you will now see tab names like "Group 1," "Group 2," and so on, clearly indicating where to click to edit your Rec Group definition. This improvement ensures a smoother and more user-friendly rule management process, helping you fine-tune your personalization strategies effortlessly.
Jira: ENG-26304
Engage
Filter and refine content by tag in Personalize API
With the Personalize API, you now have the ability to filter results using tags. This feature enables Engage to deliver content based on dynamic inputs. This enhancement simplifies content delivery, reducing the need for numerous context-specific campaigns.
For instance, consider a list page where the shopper is viewing the shoes category and has applied filters for women's shoes and the Nike brand. You want their personalized content to match all these interests (shoes + women + Nike). If no content covers all these themes, we can drop some preferences, like Nike, to find matching content (e.g., content matching shoes + women).
To implement the above example, you can use the new "tagFilter" and "tagRefinement" parameters with the Personalize API. Assuming the themes mentioned are available as tags, use "tagFilter" for mandatory tags and "tagRefinement" for optional ones. Multiple tags can be passed into both parameters using the pipe (|) as a separator.
tagFilter=shoes|women&tagRefinement=Nike
The content selection logic will follow the existing rules defined by the Priority setting for the campaigns. However, if a higher-priority campaign yields no content due to the "tagFilter" parameter, a lower-priority campaign with matching content will be used.
Note: "tagRefinement" will only be applied when "tagFilter" is being used. Currently, this feature is not available in client-side p13n_generated.js integrations.
Jira: ENG-26564
Social Proof - Add Inventory Messaging
This feature allows clients to add Inventory messaging to social proof strategies, alongside views, Add to Cart (ATC), and purchases metrics. Clients need to include inventory quantity and inventory updated date-time in their catalog feed for real-time inventory calculations.
Inventory Messaging introduces a new "Inventory" message type, enabling clients to trigger messages when inventory drops below a defined threshold (default is 10). Customizable message text and headings enhance user engagement.
Jira: ENG-23956
Data Engineering
Enhanced Reporting Menu
In this release, we've enhanced the reporting menu by adding two new reports:
Recommendation Type Analysis Report: Gain insights into recommendation types.
Sessions Report: Track user sessions conveniently.
These reports are placed in alphabetical order for easy access, improving your data analysis capabilities.
Jira: ENG-26776
Other Feature Enhancements
The following feature enhancements and upgrades have been made in the release version 23.20 during Aug 25 - Sep 07, 2023.
|
Jira # |
Module/Title |
Summary |
General Availability |
|---|---|---|---|
|
Add Data Engineering: Need external user Id as part of master data rollup |
Analytics user can now have external user Id as part of the master data so that the data source can service various clients who request custom product reports. |
07-Sep-2023 |
|
|
Dashboard: Hide InSegmentStrategies row on Model Options page |
Temporarily we are hiding InSegmentStrategies as a row on the Model Options page. This would be hidden until some front-end improvements are done. |
07-Sep-2023 |
|
|
Dashboard: Cross site request forgery phase 2 |
CORS was blocking CSRF attack request because of wrong configurations of CORS. Implemented the double submit cookie method for CSRF - Phase 2 and resolved the issue. |
07-Sep-2023 |
|
|
Recommend: Check if a Configurable Strategy with a same before saving it |
Recommend now checks if a name exists before saving Configurable Strategy duplicate strategies cannot be saved this also avoids errors in other areas of the dashboard., for example, Compound Strategies. |
07-Sep-2023 |
|
|
Science: Composite Outfit - Option to select specific products for parts of the layout |
Merchandisers can now curate outfits with precision by selecting specific products for different parts of the layout, regardless of their scores. This enhancement provides finer control over outfit creation, ensuring selected products are included in outfit generation. It streamlines the process, allowing for greater merchandising flexibility. |
|
|
|
Science: Add ‘Shop the Look’ model to Model Options |
Enabled Shop the Look model and configured categories on which the model should be built, and images are used as seed and rec images. |
07-Sep-2023 |
|
|
Platform: Ability to de-dupe the products recommended with Email |
Efficiency and consistency in email recommendations are now greatly improved. With this enhancement, emails can contain multiple layouts, each with multiple placements. Previously, when requests were processed, the same product could be recommended across different placements, resulting in redundancy. Now, the system can accept multiple placement IDs from email services in a single request. This allows the recommendation engine to eliminate duplicates across all placements within an email layout, ensuring a more seamless and effective email marketing experience. |
07-Sep-2023 |
|
|
Find > RR Server Add Channel and Region for FIND all Search requests in DataDog |
New metrics tags ‘Region’ and ‘Channel’ has been added, in cases where region and channel value is not available, used default value which is undefined.
|
20-Sep-2023 |
|
|
Find: Create new metrics Absolute Request and Error Rate in RR Server |
The Cluster Service Data has somehow got updated in QA, because if which rrserver was not able to find the search service cluster to contact the search service. After updating the cluster back to its original seeding and restarting the rrserver the issue has been fixed and we are now able to view both absolute error count and error request size matrices in DD. |
20-Sep-2023 |
Bug and Support Fixes
The following issues have been fixed in the release version 23.20 during Aug 25 - Sep 07, 2023.
|
Jira# |
Title |
Summary |
General Availability |
|---|---|---|---|
|
Enterprise Dashboard: Composite Outfit - Products are not excluding |
We have resolved an issue where excluded products were still appearing in the Outfit section of Composite Outfits. Now, products that have been correctly excluded will no longer be displayed in the Outfit section, ensuring a more accurate shopping experience. |
|
|
|
Dashboard > New Social Proof: Disabled Show/Hide options were automatically getting enabled |
The show and hide options defined as part of an experience or variation were changing automatically. The issue has been fixed now. |
07-Sep-2023 |
|
|
Dashboard: Attribute top seller configurable strategy was not displaying all attribute values |
All the attribute values are now being displayed. The issue has been fixed as the autoComplete API is being called again when the user types in a value and if the value is not available in the list, then a new attribute option is showing. |
07-Sep-2023 |
|
|
Enterprise Dashboard, MVT: Placements Context not loading in Placement MVT |
We have fixed an issue where Placement MVT was not loading for a particular client. The problem was related to a disabled placement without a name and page type. After excluding this placement from the dashboard, the Placement MVT is now functioning correctly for the client. |
07-Sep-2023 |
|
|
Enterprise Dashboard: Error while saving Affinity Configuration |
We've addressed an issue where users encountered an error when attempting to save Affinity Configuration changes. In some instances, the changes were not reflected even after receiving a successful update message. This problem has now been resolved, and user affinities are saving correctly without errors. |
07-Sep-2023 |
|
|
Data Engineering, Enterprise Dashboard: Date range is taken up different compare to input given |
Resolved an issue where the system was incorrectly interpreting date ranges, causing data for dates beyond the specified range to be included. This issue has been addressed, and date ranges now accurately reflect user inputs. |
07-Sep-2023 |
|
|
Recommend: RFM parameter for recommend is not working |
We have fixed an issue with the RFM parameter on the cart page. Previously, it wasn't functioning correctly when used with &rfb=false to filter products, causing unwanted items to be included. Now, the RFM parameter works as expected, ensuring that only products matching the specified attribute are recommended. |
07-Sep-2023 |
|
|
Recommend: TopSellers strategy returning items on top that are lower ranked in Model Browser |
Addressed an issue where the TopSellers strategy displayed products in an unexpected order, causing items that didn't meet the "Top Seller" criteria to be featured incorrectly. The issue has been fixed now. |
07-Sep-2023 |