Skip to main content

EDP Release Notes

  • Updated

Detailed information about the latest releases of the EDP. 

EDP Version 22.0 Release Notes

Core Enhancements & Fixes

Enhancement: Segmentation UI Condition Builder improvements

Cheetah platform provides a powerful segmentation system that allows customers to create Audiences using a set of rules. Audiences are used for targeting promotions, offers, communications, send eligibility, etc.

Customers build audiences by defining segmentation rules: criteria that determines if a specific member should be included in an audience. The rules can include demographic information, member preferences, machine learning calculations and other conditions, including complex combinations rules.

This enhancement adds the ability to apply multiple rule combinations for Demographics, Preferences, Machine Learning, and Calculations tabs of the segmentation rules interface, making it easier for marketers to build complex audiences.

Enhancement: Map and Transform Preview

Cheetah platform supports data import from external sources. As the source files often have different structures, the platform can transform the source data to a standard format. Customers can set rules to transform the source data and preview the expected result using the Map and Transform Preview function.

This enhancement adds the option to download the data that did not transform correctly during the Map and Transform Preview, making it easier for customers and Cheetah Technical Services to analyze errors and fix any failing transformation rules.

Enhancement: Scheduled job to sync Campaigns and Locations data

Cheetah platform utilizes the Hive database to run big data requests and analytics. This enhancement adds a scheduled job that automatically synchronizes Campaign and Location data to Hive once a day. This enables customers to run analytics on this data, and simplifies data sync configuration.

This scheduled job will be run once a day by default for new and newly updated environments. To ensure data consistency, the job cannot be deleted, but it can be disabled or rescheduled.

Enhancement: Batch Job To Sync Anonymous activities

The Cheetah platform utilizes Hive to run big data requests and analytics, and Apache Kafka to stream data (for example, member activities) to other locations, internal or external to the Cheetah Platform. Previously, whenever there was a large number of member activities in the Kafka queue, it could lead to a delay in sync for anonymous (non-member) activities.

This enhancement adds a batch job to sync anonymous activities from Kafka directly to Hive, skipping a process that doesn’t add value (the Kafka topic offsets of the Hive Activity Sync topic). This improves stability and removes delays for anonymous activities sync.

Integration Enhancements & Fixes

Enhancement: Delete via Import with multiple fields

Cheetah customers can import data from external systems, and transform it using Map & Transform rules. Transformation rules also support deleting records using import.

This enhancement/fix adds support for deleting records via import when using multiple fields in Map & Transform rules, as well as when using uppercase and mixed-case email formats.

Fix: Resolved an issue when importing External List data from SFTP

Cheetah platform provides multiple ways to integrate with external systems. One way to do this is the External Lists functionality, which allows customers to use member data imported from external systems.

This update resolves an issue that occurred during import of external data to an External List from an SFTP source.

Other Fixes, Enhancements & Hardening

  • Resolved an issue with a failing sending job when the expression has no included key name.
  • Improved handling of error messages when evaluating incorrect expressions.
  • Batch Decisioning API:
    • API return type Date is now case insensitive.
    • Fixed response for undefined or empty expression on payload.
    • Improved handling of return_data_type.
  • RTP API: improved handling of event deduplication key (cache key)
  • Fixed an issue with API service connection pooling.
  • Fixed an error that occurred when using undeployed segments.

EDP Version 16.0 Release Notes

Integration Enhancements & Fixes

Enhanced Custom Object optimization for non-custom object related columns

Background

Custom Objects are designed to extend the customer profile with object types specific to the user’s business (for example, hotel reservations). These objects contain their own set of Attributes, and can be used for segmentation, personalization, and analytics. Queries that include custom objects are processed through a custom object query optimizer to improve query performance in exports, integrations, analytics, etc.

Solution

Users can now skip optimization on Custom Object Optimizer for non-custom object-related columns. Before this enhancement, custom object query optimizers could not handle multiple tables and incorrectly applied get_json_object on the non-custom object fields.

 

Resolved an issue where campaignStat was not updating when launching a Cheetah Platform (EDP) email campaign

CampaignStat is an API endpoint used to track and display campaign status. This fix resolves an issue when the EDP campaign status did not update after sending a Cheetah Platform (EDP) email campaign.

Core Enhancements & Fixes

Support for LLAP-HA (Low Latency Analytical Processing High Availability)

Background

Cheetah utilizes LLAP (Low Latency Analytical Processing) technology for data processing.

Solution

This enhancement adds support for LLAP High Availability instances for improved stability and data safety.

When LLAP-HA is enabled, one instance operates in active mode, the other in passive (standby) mode. The passive instance serves as a backup and takes over if the active instance goes down, resulting in improved stability and availability.

 

Trigger support for Metric Expiry Activities

Background

Activities refer to some action or consumer behavior that the platform is able to capture and track, including link clicks, purchases, store visits, etc.

Activities data is aggregated into Metrics, including earned, redeemed and expired activities. Triggers are used to monitor changes to a consumer, and to execute a specific action when a specific condition is satisfied.

Solution

When an Activity expires, it will now run a Trigger for a metric expiration, which can be used to run any action that can be set up on the Triggers screen. These actions include sending an email on activity expiration, granting a reward, etc.

Member attributes: fixed an issue where the user could not modify the result value in decision table for objects

Cheetah Platform (EDP) members can have Calculated Attributes, which are dynamically calculated using their calculation rules. One type of calculation rules is a Decision Table, which is an easy way to map expected calculation results with other attribute values. When an attribute condition is met in the attributes column, the corresponding value is returned from the result column, making it easier to add multiple calculated values for different conditions. The result column may have different value types, including numeric value, object, etc.

This fix resolves an issue when a user could not modify the result value in decision table calculation rules if it was returning an object.

Member attributes: fixed internal errors in member attribute decision table

 This fix improves stability of calculated attributes interface when saving an empty decision table for example if no entries had been made at that point.

Member attributes: fixed an issue when the ‘long’ return type in the member attribute decision table could not save any value

 This fix resolves an issue when a user could not save the result value in a decisioning table when it was returning a value that had a ‘long’ numeric type.

Member attributes: fixed internal errors in member attribute decision table when adding result value for an Integer/Long type value

This fix improves stability of calculated attributes interface when saving a value of a numeric (integer/long) type in a decision table.

Other Fixes & Enhancements

  • Added a new API endpoint to delete a list of items in a custom object for a specific member.
  • Fixed an issue where a user could input a calculation result that did not match the member attribute return type.
  • Fixed member calculated attributes expression logic when using decision table column groups

EDP Version 15.0 Release Notes

Integration Enhancements

Facebook Ad Audience now Supports Lookalike Values

Users can send lookalike values to Facebook to build lookalike audiences.

Record Deletion Enhancements

Users can import a file of records that are meant to be deleted in EDP and Cheetah Loyalty.  Before this enhancement, member records would be deleted, and associated records like activity, orders, and others would be orphaned. Now all associated records are deleted.

Export Definitions Support Different Text Qualifiers

There is now a setting on export definitions that allows the user to set the text qualifier for the exported file. This setting supports the following options:

  • Double Quotes (" ")
  • Single Quotes (' ')
  • None
  • Pipe (|)
  • Custom

When the user selects "custom," then the user is able to input a character. When the user executes a file export with a text qualifier other than “None,” then the exported file wraps each field value with the selected text qualifier.

Ability to Cancel Import Jobs

Users now have the ability to cancel import jobs prior to the “inserting” job status. At “inserting” or after, the job can no longer be cancelled. Users can cancel the job by clicking a cancel icon in the Status column on the Job History tab.

Ability to Cancel ML Scoring Runs

Users can now cancel ML scoring runs prior to the “member update” status.  At “member update” or after, the ML run can no longer be cancelled. Users can cancel the ML run by clicking the cancel icon in the Action column of the Results tab.

Import and Export Enhancements

Combined the Activity Import Settings "Direct Activity Import" and "Process Earned Rule Calculations"

Cheetah has removed the setting “Direct Activity Import” from the Activity imports. This setting had redundant capabilities with the setting “Process Earned Rule Calculation.”

Place Import Skipped and Error Logs in SFTP or S3 Directory

Users can now choose to have skipped and error logs for imports placed in a SFTP or S3 Directory where the user can access them.

Setting in event imports to include event data in customer profile

When importing event data, users can now decide if that data should be included in the customer profile. All event data is accessible for reporting and analytics, but there are cases where historical event data is too old to be useful for segmentation and targeting and serves no useful purpose on the customer profile. Now there is a setting that allows users to exclude event data from profile records.

APIs for File Export Definitions

Created publicly accessible APIs for Export Definitions. The APIs support create, read, update, and delete, as well as execution of exports.

Enhanced the Error Display for Test Map & Transform Preview

There are now two tabs in the Test Map & Transform Preview: Processed and Rejected.  The Processed tab contains successful records, as well as lookup errors and other object-specific transform errors.  The rejected tab contains all records that failed generic transforms.

Imports can now process multiple files as one file across multiple import jobs

When importing multiple files as one file, Cheetah concatenates all match files into one file and imports it in one import job.

If either the number of files or the size of the files is big enough, the size of the concatenated file can exceed the total memory available for the import process. This results in a failed import. 

Users can now specify if the concatenated file should be processed across multiple jobs, and how many files should be included in each job. This allows the import to complete successfully without memory issues.

Clone HTTP API Templates

Users can now clone existing HTTP API Templates.  This will save users time and effort in creating new templates.

Expanded the List of Supported AWS S3 Regions for the S3 Connector

The S3 Connector now supports all of the following AWS S3 Regions.

  • us-east-2 - US East (Ohio)
  • us-east-1 - US East (N. Virginia)
  • us-west-1 - US West (N. California)
  • us-west-2 - US West (Oregon)
  • af-south-1 - Africa (Cape Town)
  • ap-south-1 - Asia Pacific (Mumbai)
  • ap-northeast-3 - Asia Pacific (Osaka-Local)
  • ap-northeast-2 - Asia Pacific (Seoul)
  • ap-southeast-1 - Asia Pacific (Singapore)
  • ap-southeast-2 - Asia Pacific (Sydney)
  • ap-northeast-1 - Asia Pacific (Tokyo)
  • ap-east-1 - Asia Pacific (Hong Kong)
  • ca-central-1 - Canada (Central)
  • cn-north-1 - China (Beijing)
  • cn-northwest-1 - China (Ningxia)
  • eu-central-1 - EU (Frankfurt)
  • eu-west-1 - EU (Ireland)
  • eu-west-2 - EU (London)
  • eu-west-3 - EU (Paris)
  • eu-north-1 - EU (Stockholm)
  • eu-south-1 - Europe (Milan)
  • sa-east-1 - South America (São Paulo)
  • me-south-1 - Middle East (Bahrain)
  • us-gov-west-1 - AWS GovCloud (US-West)
  • us-gov-east-1 - AWS GovCloud (US-East)

Machine Learning Enhancements

Add extra Parameters in CDS Model

All extra parameters are displayed when "Use Custom Query" is selected.

Segmentation Enhancements

EDP Segmentation available for Loyalty Eligibility

Marketers can now use EDP segments in Loyalty (e.g., Offers) eligibility as long as the segment uses a Use as Golden Record execution type. EDP segments typically compute significantly faster than Golden Record segments, so this gives marketer the ability to get a quick size of the audience before using for Loyalty.

Other Bug Fixes/Enhancements

  • Test Map & Transform Preview: Fixed a bug in import definitions where lookup transforms that did not resolve a value would cause the Test Map & Transform preview to fail.
  • S3 Buckets Across Regions: Fixed a bug that prevented EDP and Loyalty S3 Connectors from connecting to S3 buckets that were in different AWS regions than the account environment.
  • Imports: Fixed a scenario where records could be created in one database but not the others. This ensures that records are in sync across all databases.
  • EDP Segment: Resolved a bug where there are members in the segment tab.
  • EDP Segment: Fixed an issue where EDP Segment will be removed from the list when the segment is refreshed, and the member is no longer valid.
  • EDP Segment: Added an additional field in the segment metadata to show that the EDP segment is written in HBase.
  • Integration Service: Corrected a scenario where a backlog of imports can cause delays in updating import job statuses.  

Archive

Was this article helpful?
0 out of 0 found this helpful
Return to top

Related articles