Skip to main content

14.08 Release Notes (04/14/2014)



Added support for start date/end date in the Flat File Feed Format. This has the same functionality that was present in the XML feed.


Promo+Email (Beta)

User Guide

If you're interested in this feature, please contact your dedicated CSE.

Promo in email.png

Referral Targeting Enhancement

The referral targeting feature has been enhanced to now accept only valid URLs for referring sites. When a new referral site is added or deleted, it is automatically added or deleted for other campaigns (global).

RichPromo User Guide update

Referrers .png

Personalize API: Returns JSONP

New request parameter: callback

Personalize API: TrackingUrl

TrackingUrl is an additional field in the Personalize API response section. It is the the email open event tracking URL. It should only be used when Promo is rendered within an email.


Attribute Filtering

RichSort result can now filter by attributes. Consumers are now able to filter their personalized category listing by price, brand, color, and more. Filtering can be done both client-side and server-side.

Better Tuning

Retailers can now improve their personalized category listing by tuning personalization and diversity. Diversity controls how much variety of brand and category are incorporated into the results. Personalization controls how many personalized products are incorporated into the results.

better tuning.png

Consistency Control

RichSort constantly analyzes user history to personalize the category listing. While personalization is good, sometime too much personalization might be confusing to the user since the category listing might change from one time or another. Consistency control (in Site Configurations) allows retailer to set a cut-off time for user events that will be used by the RichSort algorithm to personalize the response--making the result more consistent and improving the user experience.


Build Your Own Strategy (BYOS)

For the first time ever, RichRelevance Datamesh customers will have the ability to create their own algorithms or strategies on top of the RichRelevance managed cloud Hadoop infrastructure. These strategies can be available for use in RichRecs within hours. Data scientists and analysts for retailers can use a programming language similar to SQL to create these algorithms.  These algorithms have immediate access to online click stream data that RichRelevance collects already. RichRelevance can also import other data sets that a retailer wants to use for creating algorithms. Clients can contact their Account Manager for more details on how to build their own algorithms on the RichRelevance platform.

  • Was this article helpful?