Skip to main content
RichRelevance

15.15 Release Notes (October 15, 2015)

Dashboard & API

Modification to recsForPlacements API

The recsForPlacements API was modified so that that productURL and clickTrackingURL values are in the response when the client asks for minimal data. Recommend and Discover clients that use our API, especially ones for server-side integrations, only ask for minimal data. This is done by using the parameter returnMinimalRecItemData. These clients require productURL and clickTrackingURL data in our response so that they can create SEO-friendly links.

Bug fixes

  • The Abandon Cart condition in the segment builder had incorrect default settings for the look back period. This is fixed.
  • When a new site config is published, it also needs to be published to the Open platform. This is fixed.


Recommend

Advanced Merchandising

  • Performance enhancements on the rule page load times. This is the first step of a bigger project. Our clients should see improvements in the page load times. More updates soon. 
  • Now supporting Portugal. CNET DataSource attribute and compatibility data is now available for Portugal.

Search history in UPS for search strategies

Clients using cookieless implementation are now enabled to use their personal search history captured in UPS for personalized search-based recommendation strategies. This includes all Recommend search-based strategies and the myRecs API.​

Clarification on resetting values on catalog feed

Clients had questions about how to reset values previously added to the catalog. We reviewed the code, and updated documentation to provide clarification on this process. 

Offline feed ingestion report

We incorporated offline feed ingest monitoring and reporting to Feed Herder for visibility into the data ingestion process.

Bug fix

  • The recsForPlacement API is updated to support regions and region-based pricing. A bug blocked delivery of region-based pricing, and this is fixed. There is no structural impact to an API call that specifies a region. It might provide a value in “regionPriceDescription” in a full response object where it that field is currently blank (but present).


Engage

Content copy

Copying content within Content Browser is now supported. When copying content, all tags, attributes, and attribute values are copied with it. This makes administration of content much easier for clients. Creating a new piece of content simply requires the client to copy it and change attribute values for their source URL and clickthrough URL parameters. It's a big time savings for clients who enter content directly into the dashboard.

Rule and content preview

Previewing a rule or content on a web site is now available. This is done by using a specific parameter and having the web site rule and/or content published to the integration environment. Updates are seen before production activation. This is very useful for clients that have multiple rules running at the same time on a page, or multiple contents within a rule.

With this feature, clients are able to see what shows on a page or placement when a specific rule is selected; then, within that rule, the client can force a particular piece of content to show, and then see how that content looks on their site. A URL with that specific rule and content can then be shared with others to see the same selections.


Discover

New Type Ahead search API

Type Ahead search API functionality remains the same; however, the API endpoint is new. This change allows us to support more clients across our data centers. It also means that Type Ahead API no longer shares the same infrastructure as the User Profile Service (UPS) and doesn't impact UPS performance. 

We'll train our CS team with the new API, and we plan to have all new installations of the Type Ahead feature use this new API. We'll also work with our CS team to migrate existing customers to the new API over the next several quarters.

You can find more details of the new API end point here: http://developer.richrelevance.com/apis-2/servicetypeahead/


Science and MVT

Margin Strategies can take absolute or relative input

Edit section
  • The existing Margin strategies (comparable to the "EV" strategies, but optimizing for expected profit as opposed to expected revenue) required client input via the catalog feed to be in the form of a percentage. This update makes it possible for the client to provide a revenue amount that reflects the profit earned by selling a unit of a given product, and have the Margin strategy score appropriately.
  • Was this article helpful?