Self-service predictive business insights with Amazon SageMaker for Tableau

Madeleine Corneli, Sr. Product Manager, Tableau 
Holt Calder, Data Architect, InterWorks
Dylan Tong, Global Segment Lead Architect, AI Augmented Analytics, AWS


Foresight is a competitive advantage. The ability to forecast demand and predict behavior can drive business growth, reduce churn and attrition, and optimize processes and supply chains. 

Unfortunately, many organizations lack the tools, technology, and skills to predict what’s next. At this critical inflection point, they need guidance on how to improve and strengthen AI and machine learning (ML). Nearly 90 percent of organizations have low business intelligence and analytics maturity according to Gartner, and the advanced analytics capabilities required to build predictive models have been limited to sophisticated data science teams. 

We want to change that. We want to help our Tableau and AWS customers democratize data science so that more people can benefit from the power of AI-driven analytics. Amazon SageMaker for Tableau empowers you with self-service predictive insights tailored to your business. We integrated Tableau with Amazon SageMaker so you can blend real-time predictions from AWS-managed models with Tableau visualizations.

Recent innovation in analytics and data science make it possible to integrate ML-driven insights directly into the flow of your analytics. Tableau, AWS, and Interworks have partnered to create a solution that combines the scalability of the Amazon SageMaker data science platform with powerful analytics from Tableau to deliver an end-to-end advanced analytics workflow. This solution is available as an AWS Quickstart to help you deploy quickly and easily.

Amazon SageMaker for Tableau architecture, 1) connecting Tableau through APIs managed on Amazon API Gateway, 2) authenticating API calls against user profiles managed by Amazon Cognito, 3) executing data transformations on AWS Lambda, transferring data to CSV to be used by Amazon SageMaker Autopilot, and 4) securing and running communication with model endpoints through a virtual private cloud.
Amazon SageMaker for Tableau architecture. Credit: AWS Partner Network

AI-driven analytics on AWS with Tableau & Amazon SageMaker

Dig into our blog post on how to leverage AI-driven analytics on AWS using Tableau and Amazon SageMaker. We illustrate how to turn data into predictive insights using the example of a bank’s marketing team that wants to learn from previous marketing campaigns and past customer behavior. The blog post walks through how to use the Tableau SageMaker integration  to launch a successful term-deposit campaign and accurately predict campaign performance. 

Our actionable playbook guides you through each step of the process, from problem to solution: 

  • Formulate machine learning (ML) problem
  • Explore data 
  • If needed, prep data using Amazon SageMaker Data Wrangler 
  • Train models with Amazon SageMaker Autopilot
  • Evaluate and deploy models
  • Integrate ML models with Tableau using Amazon SageMaker for Tableau
  • Generate predictive insights

Amazon SageMaker for Tableau is our first step toward empowering you with tailored, self-service predictive insights. Read our full blog post over at AWS to get started. We can’t wait to see how you use foresight to build competitive advantage for your business. 

Subscribe to our blog