New features and changes in version 8.5.0

Interact List process box added to batch flowcharts in Campaign

A new process box has been added to Campaign batch flowcharts to allow users to easily define the tables containing candidate offers to be served by the Interact Runtime server. The new process box, called Interact List, operates in a similar manner to a Call List or Mail List process box. Use the Interact List process box on a batch flowchart to determine the offers that will be served to customers by the runtime server, including the following choices:

  • Offer suppression at an individual level (a "black list")
  • Offer assignment at an individual level (a "white list," or score override)
  • Offer assignment at an audience level (global or default offers)
  • Offer assignment by custom SQL query

The runtime server has access to the output from this process when you deploy the interactive campaign. Note that a batch flowchart may contain multiple instances of the Interact List process box. (ENH10375)

Enhanced Learning (ENH10650,ENH10651,ENH10652,ENH10654)

The Interact Learning feature has been enhanced in the following ways:

  • In addition to the Global learning model already existing in Interact, you can now enable learning and customize learning attributes at the Interactive Channel, Zone, and Rule Group level. Each of those levels can have its own set of custom learning models. This feature is also referred to as "self-learning." The global settings for learning are inherited in the following order: Global, Interactive Channel, Zone, Rule Group, with each subsequent level having the option of adding to or overriding the inherited settings.

  • Learning Observation Mode.

    Previously, Interact could not collect learning statistics unless learning was specifically enabled. Beginning in this release, Learning Observation Mode allows Interact to collect learning statistics based on a pre-defined learning model (including the Global model) even when you are not using Interact Learning to arbitrate offers.

  • Self-learning Learning Reports. (ENH10653)

    A new report has been added to support the new self-learning models described above. Marketers can now run the Learning Model Report Analysis report in the Interact Design Time environment to compare performance of two learning models over a specified period of time.

Web Connector (ENH09370)

The Web Connector enables web pages to make calls to Interact for real-time offer personalization without having to implement low-level Java™ or SOAP calls to the Interact server. The Web Connector manages offer arbitration, presentation, and contact/response history through two key processes: Page Load, which serves the web page with personalized offers, and Offer Click Through, which captures offer click-through and re-directs it to the specified landing page.

On your web page at load time, embedded JavaScript™ code links to the Web Connector, which then uses the Interact API to return a personalized offer list, which is then added to your web page in the form of HTML and other markup fragments as needed. When a user clicks a link, it's passed to the Web Connector which uses Interact to determine the correct target URL to which the user is then redirected.

Message Connector (ENH10655,ENH10656,ENH10657)

The Interact Message Connector enables email (and other electronic media) to make calls to Interact for offer personalization at open-time and click-through, determining the offer arbitration and contact/response history through the <img> tags (to retrieve personalized offers for email at open) and <href> tags, which capture click-through to redirect the user to landing pages.

Offer Constraints (ENH10646,ENH10647)

The Offer Constraints feature allows organizations to limit and manage distribution of offer impressions, limiting the number of times an offer or a collection of offers can be presented over defined periods of time. For example, you might want to suppress an offer after a predefined quota of impressions (such as a certain number of impressions in one day) has been met, or to evenly distribute offer impressions over a period of time.

Offer de-duplication (ENH10649)

The offer de-duplication policy enhances the efficiency with which Interact removes duplicate offers from requests for multiple interaction points. To accomplish this, a new call has been added to the Interact API called getOffersForMultipleInteractionPoints, which retrieves a list of offers that spans a list of specified interaction points. The API call also specifies whether the Interact server should apply de-duplication to the returned list.

Performance enhancements across Interact

Numerous performance enhancements have been implemented across all of HCL® Interact, involving some of the following areas:

  • Contact history session caching and other file-based cache writing (ENH10959, DEF059773, DEF059774)
  • Duplicate response history entries in ETL queries are handled more efficiently (DEF055886)
  • Improved memory handling for Learning (DEF059772)
  • Learning aggregation in general is handled more efficiently (DEF057236)
  • OfferBySQL performance has been enhanced (DEF055126)