CX Journeys Winter 2022 Release Notes

Released starting January 2022.

Backend

Support for stored procedures

To extend programmatic capabilities, the ability to execute a stored procedure written in SQL has been added. A procedure is created in BigQuery database by your PS team as follows:

Execute a stored procedure in SQL.

The following procedure can now be executed from the CQL Editor:

Call procedure from CQL.

To view the result in the database, enter the following:

Query data table.

Creating a churn predictive model out of the box

Journey Analytics provides an out-of-the-box Churn Prediction model that is based on customer behavior in the last 90 days.

To build a model in your project, select the option "BQML" under the Add Model menu.

Add Model menu.

Then select the CHURN model from the list of Template Models.

CHURN model.

Within a few minutes, the system creates a table called CHURN_TRAIN_TABLE in the schema and your models list will contain the new churn Model.

To use the model for churn prediction, use the following syntax (modify the date_range and the filters in line 43):

Prediction syntax.

The results tab should display the prediction (churn/return) for each customer as well as the probabilities of that classification.

predicted_churnedpredicted_churned_probsuser_id
returned[{v={f=[{v=returned}, {v=0.54758768975067418}]}}, {v={f=[{v=churn}, {v=0.45241231024932582}]}}]6677906950825497643
returned[{v={f=[{v=returned}, {v=0.75351409543443248}]}}, {v={f=[{v=churn}, {v=0.24648590456556752}]}}]5758572130776445936
churned[{v={f=[{v=returned}, {v=0.031422519444205883}]}}, {v={f=[{v=churn}, {v=0.96857748055579407}]}}]6181753117297146221

In the above example, the first column predicted_churned contains the predicted behavior of our customer. The predicted_churned_probs column contains the probabilities for the model prediction.

Adding a new BQML Model to predict results

Similar to the above example, you can add a new Big Query Machine Learning (BQML) model in your database. Work with your PS team to update the model in your database.

You can register and use your BQML code using the "Add Model" feature.

Our team can help you to build LTV and other predictions models directly in your database.

IP address of the NAT

NAT servers have been upgraded to enhance security.

Ensure that your firewall accepts connection requests from Medallia IPs 35.199.16.189, 34.86.159.106 when adding a new linked data source.

New UI capabilities and enhanced report builders

New report visualization: Bubbles

Bubble chart visualization aims to look at relationships between three numeric variables.

Each dot in a bubble chart corresponds to a single data point, and the variables' values for each point are indicated by horizontal position, vertical position, and dot size.

Each dot is assigned a label or category (either displayed alongside or on a legend). Each plotted dot represents a third variable by the area of its circle.

Colors can also be used to distinguish between categories or used to represent an additional data variable.

To select the Bubble chart visualization, select the Bubble option in the reports visualization menu.

Bubble option.

You can also define a conditional color formatting based on the third column value.

Conditional color formatting.

New report visualization: Sparkline

Sparkline Chart (or simply Sparklines) is a tiny-sized chart, causing (or being made possible by) the absence of axes or any notes, which makes it impossible for viewers to see precise values. Sparkline Charts are great at showing the big picture.

There are three types of Sparklines:

  • Line Sparklines, the simplest and most common type of a Sparkline Chart that consists of points connected by line segments.
  • Area Sparklines, a Line Sparklines graph in which the area between the series and the X axis is filled with color.
  • Column Sparklines, a Sparklines-based data visualization where each point is depicted as a vertical rectangle, the height of which corresponds to the data value it represents.

Sparkline.

Sparkline.

When selecting line/area/bar chart, you can switch on the "Spark Line" toggle and then select the aggregation Function (AVG/SUM…) , the font size, and the text color of the big number.

Report visualization option in the reports list

In the reports list, an icon now appears that describe the visualization of that report.

Reports list.

Quick edit for reports on the dashboard

To edit the report settings (standalone or in a dashboard), select the Quick edit icon in the report header.

Quick edit.

The settings menu will appear where you can modify the existing report's settings.

You can choose to save the modified settings or exit without saving changes.

Settings.

Change visualization of reports in dashboard

To modify/change report visualization (standalone or in a dashboard), select the Visualizations icon in the report header.

Visualizations.

The visualization menu appears appear where you can modify the existing report's visualization.

You can choose to save the modified settings or exit without saving changes.

Visualizations.

Visualizations.

Journey analysis based on a virtual property

In the path builder, you can now create a path based on a virtual property (expression).

To do this, first create a new virtual property based on any CQL function.

New expression.

Then select that virtual property in the path builder widget and show the journey of customers across that property.

Widget.

Dashboard titles

Long names are now enabled for dashboard titles.

Renamed Data Sources page

Data Sources page.

Clone Project's dashboards/reports

Clone Project's dashboards and reports also clone the drill through the relative report.

Event trackers

SDK

A new version of iOS and Android SDK has ben released.

Replacing Android SDK

This is the "aar" link.

Update As a Gradle dependency:

implementation 'com.cooladata.android:android-sdk:2.4.2' 

Update As a Maven dependency:

<dependency>

  <groupId>com.cooladata.android</groupId>

  <artifactId>android-sdk</artifactId>

  <version>2.4.2</version>

  <type>aar</type>

</dependency>

Instructions

Download aar and copy it into the "libs" folder in your Android project.

Validate that you have the OkHttp library on your classpath as well.

Replacing iOS SDK

The new version is located here. Download the latest iOS SDK.

Update your  https://docs.cooladata.com/ios/#Step_2_8211Initialization 
platform :ios, '9.0'
use_frameworks!

target 'MyApp' do
pod 'CoolaPod', :git => 'https://github.com/cooladata/pod.git';
end