Service Updates & Release Notes

Learn about new features and bug fixes for Kibo products.

Real-Time Individualization

October 30, 2018 — Service Update

Features

  • Inspired by Purchases Strategy: This new strategy will recommend complementary products based specifically on a user’s past purchase. While all Kibo RTI strategies leverage purchase history in their data modeling, this strategy is reacting specifically to an item the consumer bought and recommends products they might want to buy in addition to their recent purchase. This strategy can be used anywhere throughout the customer journey, but would be most compelling on a “My Account” page or in order related emails, such as order and shipping confirmations. Please contact your Customer Success or Professional Services representative to enable this new strategy.
  • Any/All Prefix in Content Groups: Users can now use a prefix of “Any” or “All” when building Content Groups. This feature allows you to write expressions with more flexibility to create groups of products based on a collection of attributes.

Previous Updates

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
NAAll pages in VUE will load faster with new caching improvements.

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
NAAmpersands in the email code generation passThru parameter are now being correctly encoded.
NAText attributes containing Japanese characters are now rendered in order by the Image Server.
NAGuide render image quality no longer becomes noticeably pixelated when reducing image size by more than 50%.

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
NAStrings with Katakana characters now wrap correctly for rendered email images.
NAImageserver render API will no longer use the default image style if an imgStyle parameter is supplied in the render request. This fixes a bug where an unexpected or duplicate image could be rendered in email recommendations.

Features

  • Template Routing Expression Updates: Merchandisers in power mode can now use ANY and ALL in a routing expression so that recommendations can be routed based on a collection of attributes.
  • Timebox Configuration Updates: A time box sets a time limit for recommendations to be evaluated and returned. Separate time values can now be defined and modified for both RTI email and web recommendations. If considering adding separate timebox configurations to both RTI email and web recommendations, please contact a customer success representative.

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
NADaily, weekly, and monthly reports are now automatically regenerated when the previous day has partially missing data.

Features

  • Multiple Boost/Bury Expressions: Merchandisers now have the ability to apply multiple boost/bury expressions to a single widget. The UI for widget boost/bury has been updated in the VUE admin.
  • Timebox Configurations: A time box sets a time limit for recommendations to be evaluated and returned. Unique time values can now defined for email and web recommendations. If considering adding separate timebox configurations to both RTI email and web recommendations, please contact a customer success representative.

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
NAFixed the implementation of custom fallback images for email recommendations. A URL can be defined in the image server API to implement a custom fallback image for email recommendations. Please reference Kibo’s API Documentation for further details.

Features

  • Titles for Ensembles: Ensembles now have titles that reflect the product name of the primary product in the ensemble.
  • Improved Ensembles Scrolling: When scrolling down in Ensembles, rows will now load twenty at a time to reduce page load times. This will allow clients to create as many ensembles as they desire.
  • Ensembles UI Changes: Minor UI changes to improve the look and feel of ensembles.
  • Widget Reordering: Users can now reorder widget priority on the page template without having to delete and recreate the widget in the desired order. The ability to reorder widgets allows the user to quickly reprioritize them on a page template, to ensure that a pinned or ensemble item is returned and not affected by our deduplication process.
  • A/B Test Enhancement: The test metric is now visible on the main A/B test page. The user no longer has to navigate to the individual A/B tests to see which test metric was chosen.
  • API Documentation: We have released our Real Time Individualization API documentation to the public. They can be found here.

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
NAFixed an issue where some Japanese characters would return a question mark when used in an A/B test name.

Features

  • Ensembles (Beta): Ensembles is now available for beta testing. With this new, self-service UI, RTI customers can now create custom product recommendations to display throughout the consumer’s journey. This new feature will give clients the flexibility to merchandise collections of products, making it easier for the consumer to “complete the set.” Clients will see a new tab in the RTI Admin for Ensembles, which is deactivated for all users by default. If you are interested in participating in our beta test, please contact a Customer Success representative to see if you meet the requirements for Ensembles to be activated!

Features

  • Japanese Internationalization: RTI now provides the ability to change the language of the Admin interface to Japanese. Japanese customers will have an easier time navigating the Admin interface, which allows them to have a better understanding of the changes they could make to better customize RTI to their needs.
  • Email Recommendation Enhancements: RTI has enhanced the ability to use product IDs as context in email recommendations. These enhancements allow RTI to provide more consistent recommendations when a product ID is used as context instead of a product URL.

Features

  • Announcements: The most recent date your catalog was successfully updated is now visible on the RTI Dashboard. Previously, whether your catalog successfully updated was tedious to find. This information can now be found when you log into RTI VUE, in the announcements box to the top right of the dashboard.

Features

  • Strategy Name Change: “User Specified” is now called “Ensembles”. This was done to reduce confusion when trying to select a widget strategy to leverage “Ensembles”. The Ensembles strategy provides clients with the ability to select and attach custom recommendations to specific products. This allows the client to display matching pieces to complete the outfit or set for the shopper to see as they browse those products.

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
NAFixed an issue where, after successfully logging in and hitting the back button in the browser, the user would receive a 403 error message when trying to login again.

Features

  • Overlaps Operator: A new operator for Filters and Content Groups that works with categorical data. Works as an intersection for attributes with multiple values, needing only one matching attribute value for a product to be included in the results.

Features

  • Ensembles—RTI now supports the ability to import customized contextual recommendations, also known as “Ensembles,” for customers’ specific business needs. A new widget Strategy has been introduced that leverages context specific recommendations provided by the customer. These are provided via a file similar to the catalog updates and the system will process these and make them available.
  • Improved Latent Preference Model—Optimized performance of the Latent Preference Personalized (AKA Latent Preference Model) Strategy for customers with very large catalogs.

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
6217Fixed an issue by removing the expressions "Within" and "Outside" in the content group rule builder, as these operators were not relevant to content groups and causing errors to occur.
12368Fixed an issue that caused the user store to use excessive amounts of resources and a lot of time.

Features

  • Rebranded RTI Admin Console—RTI pages are now rebranded with Kibo colors and an updated user interface for the login page, Admin console, and the Contact Support page. All functionality remains the same.
  • Arithmetic Operators to Filter by Range—Merchandisers now have the ability to create filters by range by leveraging percentages and other arithmetic operators. Arithmetic operators are added to the contextual help only for numeric attributes.
  • Attribute Origin on User Details Page—The User Details page now displays a column that shows the origin of the user attributes.
  • User Trail Recommender—You can now individualize results even further with the new User Trail Recommender by producing recommendations that are more specific to the interests the user expressed throughout their journey on your site. The User Trail Recommender analyzes the full user trail in order to generate recommendations based on how much to weight or decay items in the trail depending on various factors.

Bug Fixes

The following list summarizes enhancements and resolved issues for this release:

Support TicketResolution
6061Fixes an issue that prevented the user store from recording the external ID for some customers.
6059Fixes an issue that prevented optimal search results from displaying.
6072Fixes an issue that caused UserStoreCopyToSchema20Job to fail for NumberFormatException.
5886Fixes an issue that prevented document details pages from populating correctly.
Angela FausoneRelease Notes – RTI