AutoML, ABTs & Visualization for Touchpoint Performance with SAS Customer Intelligence 360

[#1]

No matter what you call them, every analyst is guilty of making the following statement to their leadership team.

"I spend more than 80% of my time preparing data, and less than 20% actually performing analysis."

Speed bumps like this usually emerge when marketing teams require advanced insights like propensity scoring, algorithmic segmentation or next-best-actions. For example, have you ever tried to extract HIT (or click-level) data from your preferred marketing cloud vendor? It's not formatted for machine learning or AI applications, and time is lost in the complex efforts to re-engineer that information.

Accelerating through this challenge has been a key area of interest at SAS, and the solution is referred to as analytic base tables (or ABTs). It is the automated process of organizing data into a flat table schema that's used by analysts for building analytical models and scoring (predicting) the future behavior of a customer. A single record in this table represents the subject of the prediction (such as a customer or anonymous visitor) and stores all data (variables, features or predictors) describing this subject.

ABTs are now available as part of the structured data model available to users of SAS Customer Intelligence 360 for a variety of customizable analysis purposes.

Figure 1: ABTs for AutoML & insight acceleration

Let's walk through an example for touchpoint performance measurement.

The Attribution ABT is the table that SAS Customer Intelligence 360 uses as a source for attribution modeling. Each row in the attribution table represents one customer interaction. The table shows three types of data:

1. Origination data (traffic sources)
2. Tasks (the targeted tactics a brand takes when a user is interacting with your organization)
3. Conversion events (based on your defined macro- or micro-goals)

Analysts can download the table and take one of these actions:

• Run the data against your own analytical models in SAS, open source, etc.
• Append any other data you have, such as direct marketing contact and response records, to the table for a more robust set of information.
• Review the data that SAS Customer Intelligence 360 uses as a source for automated attribution modeling.

Accessing this data is critical for users, and SAS has added these download programs in GitHub so that you can subscribe to notifications for program updates. The SAS software version of the download program can be found at here, while the Python version is here. For the latest details related to SAS Customer Intelligence 360 version releases, please visit this SAS Communities posting.

With the advances of a wider and deeper set of first-party data now available from SAS Customer Intelligence 360, the importance of actionable decisions derived from analytically-derived insights is increasing. Let's transition to the topic of recent advances in automated machine learning.

 AutoML

Automated machine learning (commonly referred to as AutoML) involves automating the tasks that are required for building a predictive model based on machine learning algorithms. These tasks include data cleansing, feature engineering, variable importance, model selection and hyperparameter tuning, which can be tedious to perform manually. Platforms that provide this capability offer many benefits, such as empowering analysts by giving them a start at a machine learning workflow, as well as allowing advanced data scientists to spend less time on model design and more time on making an AI-enhanced marketing campaign a reality.

Automation is not intended to replace the role of data scientists; ideally, there should be support for intervention in these systems to allow the performance of tasks such as domain-specific feature engineering, which can be a critical component of improving the performance of predictive modeling. These systems should be transparent with regard to the algorithms being used, so that users can be aware of, understand, and trust the insights being generated.

SAS provides different levels of automation that can be included in the machine learning pipeline-building process. Users can do any combination of automated tasks, such as having the system determine variable roles and levels, create the best transformation for numeric features, generate new features, and more. Alternatively, the entire process can be automated, through a graphical user interface as well as using a REST API. As an example, the Machine Learning Pipeline Automation API can be integrated into your own applications to automatically build a pipeline, run it, and return the champion model, which can then be deployed.

With that said, I invite you to view a video and technology demonstration that will address the following topics within SAS Customer Intelligence 360:

1. What is marketing attribution for performance measurement?
2. What are the different ways of analyzing this business problem?
3. What is the CI 360 Attribution ABT?
4. What is AutoML, and how does SAS Visual Data Mining & Machine Learning support SAS Customer Intelligence 360 users interested in this capability?

Learn more about how the SAS platform can be applied for marketing data management here.