Q4-2024 Release

Features/UI

• If you build a segment using base segments (rather than profile properties), any inverse segment you create from it now works as expected and shows the correct number of profiles. With this fix, you can continue to use inverse segments to track lifecycle stages.

• When building a segment using a profile property with the Text data type, selecting the "has a value" operator now works as expected with all values properly displayed.

• When building a segment using a profile property with the Decimal data type, the graph now loads successfully once you set a range of values.

• When you manually refresh a timeline event rollup (from the right sidebar), a new event no longer gets added to the activity stream on the platform homepage.

• Profiles that share a unique identifier used for a merge rule will now merge, as expected, if they are within the same domain group.

• When defining completion criteria for a lifecycle, you can again access tooltips for the checkboxes "Select lifecycle completion event" and "Allow profiles to reenter the lifecycle."

• An issue was fixed on the Privacy page that disabled "Europe (GDPR)" from the list of legislation zones and prevented you from checking it.

• The Journey Recorder feature, which had limited customer adoption, was removed from the BlueConic Simulator to provide a simpler UI.

• The "Don't show on Segments tab" setting is now unchecked by default for three read-only, system-created profile properties so you can use those properties immediately to build segments:

  • Creation date

  • Last modified date

  • System has property

• To ensure consistent property values for both profiles and groups, you can no longer manually alter the value for the Creation date property or create additional values. If you currently have multiple dates for that property, you will only be able to export the first listed date moving forward.

• On the Profiles page, you can enter an extremely long value (i.e., more than 500 characters) in the "On unique identifier" search field without UI issues occurring.

• When creating a Send Email to Visitor Dialogue, apostrophes and quotation marks added to the email subject line now display as expected.

• When viewing an item from a table in BlueConic, such as a profile from the Profiles page, clicking the left or right arrow key on your keyboard opens the item that appears directly before or after it in that table.

• When sorting by Type on the Profile Properties page, properties with the Group ID type are now sorted correctly.

Connections

• Improvements were made to the way the SFTP Connection exports and downloads large files, resulting in better run times. Additionally, error reporting was enhanced to provide more details, enabling faster troubleshooting of configuration issues.

• Updates were made to the LinkedIn Connection that resolved access token expiration error messages from occurring.

• A fix was made to the Amazon Web Services (S3), Snowflake, and Braze connections that prevents unwanted authentication errors from being thrown.

• The Klaviyo Connection was updated to the latest API version, resolving incorrect export goal errors from being thrown.

• Cleanup now works successfully in the Adestra Connection when multiple goals are enabled or when a single goal has two or more lists selected.

• The Salesforce Marketing Cloud Connection no longer has version 2.24.0 incompatibility errors that were causing batch import goal matching issues.

• Optimizations were made to the Shopify Connection to handle large data imports more efficiently and avoid bulk operation timeouts. Additionally, the retry mechanism in the Shopify Connection was adjusted, resolving internal server errors. Both of these enhancements ensure smoother synchronization between both systems, even when substantial data volumes are involved.

• Error reporting was improved in the Snowflake Connection, now capturing and communicating IP-related authentication failures. Run logs accurately reflect IP blocking events, enhancing visibility and enabling faster troubleshooting of connection issues.

• An optimization was made to the Kafka Connection and its memory management, ensuring more efficient and stable connection runs, particularly for long-running or resource-intensive operations.

• An issue was resolved in the Webhook Connection where inverted section tags in Mustache templates weren't functioning correctly. This fix ensures proper conditional rendering based on property values, particularly benefiting scenarios involving identifier fields without a value set yet. Users can now reliably implement fallback logic in their request configurations, improving data handling flexibility and accuracy.

• A data collection issue in the Product Collector (SFTP) Connection was resolved that impacted Single Page Application (SPA) environments. Now, multiple product views are executed and logged correctly, ensuring accurate tracking of product view events for SPA environments.

• The Sailthru Connection can now handle empty values in matching criteria, preventing unintended merges caused by hashing empty strings. This refinement improves data integrity by only hashing non-empty values, resulting in more accurate profile matching and reduced risk of erroneous data management.

REST API

• A fix was made to segment refinements to ensure they always work as an implicit AND (instead of an implicit OR) over the base segment and thus return the correct results.

• Group refinements will return correct results with either uppercase or lowercase query values.

• When calling /profiles/{profileid}, a profile.segments expand option was added that, when used, returns both the segment name and ID. (If this option isn't used, only the ID is returned.) This update is now reflected in the Get one profile section of the REST API v2 documentation site.

• The Recommendations section of the REST API v2 documentation site was enhanced to include clearer specifications and examples of the filter options that are supported.

• compositionStrategy is no longer included in the Profile Properties section of the REST API v2 documentation site.

• The REST API v2 documentation site was updated to ensure that example API calls are formatted properly.

Python API

• In the Python API, "replaces" and "replacedBy" properties were added to the Profile domain object when requested as an include option (in a similar manner to Segments and Objectives). This allows you to analyze profile merges through AI Workbench without having to call the REST API directly.

• For domain objects in the Python API—specifically connections, dialogues, listeners, profile properties, segments, lifecycles, content stores, and timeline event types—the following properties are now available: 

  • Description

  • Creation date

  • Last modified date

  • Username of the creator

  • Username of the last modified user

  • Labels of the object

  • Status (only for domain objects that have a status toggle)

Q4-2024 Release

The Iterable Connection, which was previously available just to U.S.-based customers, is now available to customers in Europe, as the connection recently integrated with Iterable's European data center. This allows a wider range of BlueConic customers to adopt the connection to enable their customer engagement use cases.

With this update, a new "Project location" dropdown was added to the connection's Set up and run page (under Authentication) to select your region: United States (US) or Europe (EU). Selecting the Europe (EU) option provides the necessary access for European customers to import and export data between Iterable and BlueConic.

Q4-2024 Release

To enable you to create and manage custom audiences in Amazon Ads directly from BlueConic, a new "Target region" setting was added to the Amazon Ads Connection (on the Set up and run page) to select the region where your Amazon Ads account operates. Two options are available for selection: Europe (EU) and North America (NA).

Additionally, within the export goal itself, a new step was included for selecting the country of origin for your exported custom audiences to comply with the privacy policies of Amazon Ads. You have the option of selecting the country of origin by using the dropdown menu or by selecting the BlueConic profile property where this information is stored.

Q4-2024 Release

As part of the rollout for Real-Time Sync for Snowflake, a new real-time import goal is now available in the Snowflake Connection, allowing you to retrieve data in real time from Snowflake and facilitate dynamic personalization and decisioning in BlueConic.

This new goal, labeled "Import data into BlueConic - Real time," consists of three steps (which are outlined in the Creating real-time import goals section of our Snowflake Connection article). 

Upon saving the goal, Snowflake data is imported automatically into BlueConic in real time. The profile will continue to house the properties and the continually updated property values put into place by this connection goal so you can use them in segmentation, decisioning, and personalization. You can also push this data to other marketing systems based on customer engagement.

Defining channels

Through a new "Channels" setting on the connection's Set up and run page, you can select specific channels on which you want the Snowflake real-time import to be active. This allows you to control where and how real-time data is utilized within your interactions.

Q4-2024 Release

Since last quarter, several fixes have been made to the BlueConic SDKs, as outlined in the Q3-2024 BlueConic Release Notes. These are the most current versions (as of October 18, 2024):

• BlueConic SDK for Android: v5.2.6

• BlueConic SDK for iOS: v3.3.0

• BlueConic SDK for React Native: v3.2.4

For the latest updates, review the articles the BlueConic SDK for Android Release Notes, the BlueConic SDK for iOS Release Notes, and the BlueConic SDK for React Native Release Notes.

Using BlueConic and not yet adopted our SDKs?

Get started in minutes with our Android and iOS SDKs or React Native SDK to start capturing relevant data to orchestrate resonant experiences that deliver a higher return on data across all your customer touchpoints.

Q4-2024 Release

With a commitment to data security and privacy, we always look for ways—big or small—to provide additional protection for customers and ensure smoother integrations with other systems. With this release, we flagged these client-side JavaScript cookies as Secure to better protect your users' data while they're on your HTTPS website:

• BCSessionID

• BCRefusedObjectives

• BCRevision_ext_

• BCTempID

In particular, this update allows you to avoid any blockers that might occur from having cookies not yet flagged as Secure, especially when you integrate with a CMS that has security requirements for the SessionID cookie.

Note: As a fallback, the cookies BCSessionID and BCRefusedObjectives are unsecure on HTTP websites.

Q4-2024 Release

You can now use the BlueConic REST API v2 and Python API to export group data to third-party systems based on specific BlueConic segment definitions. When a segment is evaluated and profiles are returned, the matching group data is included in separate sections of the API response, similar to how Segments are handled. This allows you to access both profile and associated group information efficiently.

In the REST API, this is achieved using the expand filter. The following options were added as part of this release:

• Use profiles.profile.groups to include all group information associated with profiles.

• Use profiles.profile.matchingGroups to include only the groups that match the segment criteria.

In the Python API, this functionality is provided through two new parameters: groups and matching_groups.

This enhancement removes the dependency on private APIs or external platforms (e.g., Salesforce Marketing Cloud) to configure this type of filtering. It improves platform flexibility and efficiency, allowing you to manage your group data more effectively for precise targeting. Additionally, it helps you execute highly personalized use cases based on users' group affiliations.

Note: This Groups API export functionality is now available through export goals in the SFTP Connection, as outlined in this entry below.

Seeing a sample API response

In this example, group data is returned in separate sections (matchingGroups and groups), allowing for a clear and organized representation of the data.

{
  "id": "41ebc321-b02b-4e8a-a368-82995c81fb18",
  "creationDate": "2023-03-27T10:36:11.990Z",
  "privacyLegislation": "GDPR",
  "consentedObjectives": ["objective1"],
  "refusedObjectives": ["objective2"],
  "properties": [
    {"id": "all_companies", "values": ["company_A", "company_B"]},
    {"id": "favorite_company", "values": ["company_B"]},
    {"id": "visitclicks", "values": ["1"]}
  ],
  "segments": [{"id": "bdd64355-2d45-411b-a5ea-0bcf87318f35", "name": "Export test segment"}],
  "matchingGroups": [
    {"profilePropertyId": "all_companies", "groupIds": ["company_A"]},
    {"profilePropertyId": "favorite_company", "groupIds": ["company_B"]}
  ],
  "groups": [
    {
      "id": "company_A",
      "groupTypeId": "company",
      "creationDate": "2023-03-27T10:36:11.990Z",
      "lastModifiedDate": "2024-01-10T12:52:44.816Z",
      "properties": [
        {"id": "city", "values": ["Nijmegen"]},
        {"id": "employees", "values": ["50"]}
      ]
    },
    {
      "id": "company_B",
      "groupTypeId": "company",
      "creationDate": "2023-03-27T10:36:11.990Z",
      "lastModifiedDate": "2024-01-10T12:52:44.816Z",
      "properties": [
        {"id": "city", "values": ["Boston"]},
        {"id": "employees", "values": ["100"]}
      ]
    }
  ]
}

Q4-2024 Release

To provide significant security and usability improvements—such as real-time collaboration features and a step-through debugger—several important updates were made to enhance AI Workbench: 

• Upgrade to support Jupyter Notebook 7

• Upgrade from Python 3.9 to Python 3.11

• Updates to supported Python library packages

Existing AI Workbench notebooks that have run in the 2024 calendar year will function as they did before but with an improved notebook user interface and enhanced security and usability. Python developers will benefit from better error highlighting and enhanced code navigation features.

Understanding Jupyter Notebook 7 improvements

Jupyter Notebook 7 brings the following enhancements for Python developers:

• Simplified and more intuitive user interface

• Improved support for keyboard navigation and screen readers

• New menu options for editing, viewing, and running Python code in AI Workbench notebooks

• Additional Markdown rendering options available in the Markdown menu, including better rendering of tables, code blocks, and LaTeX

Updating AI Workbench notebooks

This release of AI Workbench may require changes to your notebook code to upgrade the Python library packages called in your notebooks, if the notebook in question has not been run in the 2024 calendar year. Review the article Updating AI Workbench notebooks for details on code library dependencies and steps you may need to take, particularly these sections:

• How to check for required Python library package updates

• List of required Python package version updates

• Notes on removing the ipywidgets and python-dateutil library packages from the pip install

• Updating your standard and custom AI notebooks

• FAQ on updating Python library packages for this release

Q4-2024 Release

To provide more clarity and real-time feedback when building segments and enhance the user experience overall for segmentation, a variety of updates were made to Segment condition pages:

• A new "View profile counts" setting was added that enables you to load property values with or without their profile counts. You can turn this setting on or off at any time to switch between a detailed view with profile counts and a streamlined view with only values, depending on what suits your workflow best.

  • When this setting is enabled (the default), both property values and their associated profile counts are displayed, giving you detailed insights into how many profiles match each value. This allows you to make informed decisions based on the size of each segment.

  • When this setting is disabled, only property values are displayed, which speeds up the data retrieval process. This allows you to quickly create and adjust your segmentation logic without waiting for count data to load. (If you're working with large datasets or require faster loading times, you might choose to uncheck this setting.)

• For large tenants with vast amounts of data, you can now configure a segment without having to wait until values and counts are loaded. (The loading icons that display no longer block the configuration of the segment.)

• The OR-block counts on the left (i.e., the filters separated by the OR condition) and the segment profile count at the top right are now loaded after two seconds of inactivity to improve system performance.

• When you select a property with a large number of values, those values now load progressively in blocks of 40. The first 40 values are shown immediately, and another 40 appear each time you scroll down to the end of that block.

• To alert you that no more than 1,000 values can be loaded on the page, a message now appears when you scroll down to the 1,000th value: "Limit of 1000 values reached."

Additional UI updates

• If you return to the list of values after your segment condition has already been configured, those configured values are now loaded immediately before the other values (and before all profile counts).

• The "Filter by another value" option was relabeled "Add value" for better readability; it was also moved to the top of the list of values, preventing it from being moved down as more values load.

• The "Select all…" option now displays at the very top of the list of values (with no horizontal line separating it). It is loaded before the other values but remains disabled (and does not show an exact count) until loading is complete.

  • Note: If there is a large number of values, this option is labeled "Select first XX values"—listing the current number of values after each block of 40 is loaded.

• The "Search value" field now remains visible at all times and can be used to search even before all values are loaded.

• For properties that use graphs (e.g., bar charts, line charts), a skeleton loader now displays while the graph is loading. You can still configure the conditions during this loading time.