Traditional Customer Segmentation techniques involve the use of common customer features like customer income, demographics, life stage, attitude, and behavior to help segment target audiences. According to a recent IBM report, these features make up just 20% of data and the other “80 percent of data comes from previously untapped, unstructured information from the web such as imagery, social media channels, news feeds, emails, journals, blogs, images, sounds and videos. Relying only on a traditional segmentation approach misses the mark on garnering high-quality marketing opportunities. The limited set of categories limits the volume of actionable insights a company can take to better serve their clients.

Likewise, companies lose out on additional avenues of personalization by limiting the customer to one segment. In reality, customers placed in separate segments may still share common attributes. Valuable marketing opportunities lie not in the segments themselves, but in their attribute overlap.

For instance, think of two client segments. Segment 1 may include unmarried, home-owning, degree-holding women under 45 years old. Segment 2 may include middle-income, married parents of high-school-age-children in a particular school district. There can be a common attribute present in both segments like shopping in the same specialty store, for example. This attribute may also show up in a third segment of minority millennials starting their own digital business. At first glance, you wouldn’t think these three customer segments would have nearly anything in common – until you take a closer look.

Seeing such an overlap between client segments allows you to:

• Boost revenue for low-value clients
• Formulate strategies to grow and maintain high-value cohorts
• Finding a new pocket of clients
• Designing an affinity program for clients with a more personalized appeal

The machine learning model uncovers customer interest patterns that are otherwise not easily recognizable, allowing these new offerings to be promoted across several customer accounts. With this accelerator and Db2 you now have deeper insights to better serve your clients, allowing business intelligence (BI) analysts to use their everyday tools such as SQL query in Db2 to discover predictions from the in-database deployed AI models.

Machine learning (ML) pipelines are currently managed by data scientists who have knowledge about the platforms and languages needed to run machine learning (ML) algorithms. Current BI tools don’t support involved ML algorithms. Because of this, executive insight, dashboard view, and report creations are cumbersome with complex machine learning outputs.

What if allow the data scientist to act as the backend developer? This would enable them to build this entire machine learning pipeline externally. In a Jupyter notebook in Watson Studio, for instance, the data scientist can deploy the code into the database and have the business analyst use SQL to execute that machine learning pipeline. This solution would solve not only operational problems, but also difficulties associated with machine learning execution.

With the push of a button, an industry accelerator can become available to a business analyst without needing to write any Python code. Furthermore, with a few SQL queries, a BI would use the model predictions to take action on insights found using the Customer Segmentation accelerator.

As the image shows below, the also accelerator allows you to access value beyond customer segmentation like increasing response rates in marketing campaigns due to visibility of customer interest patterns. We can only tap into this by first knowing more about our customers with the customer segmentation use case.

Industry Accelerators on IBM Cloud Pak for Data provide tools to help you shorten time-to-value from demonstration to implementation. Learn how these accelerators can help you expedite your business strategy by exploring the new Accelerator Catalog.

For help getting started on your data science project, let our experts assist you. The IBM Data Science and AI Elite (DSE) team works side by side with your team to co-engineer AI solutions and help your business prove value at no cost. Get the skills, methods and tools you need to overcome AI adoption and to solve your business challenges quickly.

Explore the IBM Accelerator Catalog.
Request a consultation to get started.
Learn more about the IBM Data Science and AI Elite.