THINK Marketing

A marketer’s guide to customer segmentation

By , May 4, 2017

We all know how difficult it is to find the perfect gift for someone you know. So, consider being tasked with finding the perfect gift for a total stranger. Seems almost impossible. What about for 1,000 total strangers?

This is the challenge that marketers face every single day. Every day, they strive to make personalized connections with thousands or even millions of customers. And, once these connections are made, marketers are expected to maintain and grow these relationships. Customers are complicated and ever-changing. A customer may respond positively to a campaign for platform sneakers, and then be completely disinterested in shoes for the rest of the year. To form and deepen these bonds with customers, marketers need to understand and predict the changing individuality of each customer.

Analytics segmentation: A five-part primer

Choose the right segmentation strategy before you start segmenting your audience. We put together this five-part primer to help you decide where to start. When you are ready to let predictive customer behavior drive your segmentation strategy, view our smart paper.


3 steps for customer segmentation

While the ability to read minds is not a service offered on IBM Marketplace (yet), customer segmentation tools come pretty close. These tools deeply examine and understand customer behaviors, and use that knowledge to find common ground between customers. Segmentation gives marketers the opportunities to target customers based on their specific wants and needs. In three simple steps, customer segmentation can help marketers view customers as individuals, so they can maintain and deepen relationships at massive speed and scale.

Step 1: Gather the right data

Between smartphones and the Internet of Things, our digital footprint is growing astronomically. The number of customer data touchpoints has exploded. Yet, even now, only 25% of marketers are collecting customer data.

It’s difficult to analyze data that you don’t have. And, even when it is collected, the data is normally disorganized and in overwhelming amounts. To segment effectively, you need to see the whole picture. Have a system in place that captures and organizes data from multiple different channels and touchpoints. Track customer journeys across your webpages to get a sense of your customers’ web presence. And measure transactional data to understand your customers’ buying habits and preferences. Every different data point is a piece of the puzzle – the more puzzle pieces you collect, the better the view of the final picture. The more sophisticated and organized your data is, the easier it is to run analytics.

Step 2: Use customer segmentation to interpret customer behavior and draw conclusions

Once you’ve gathered your data, it’s time to put it to work. Companies are heavily investing in this area. In fact, marketing analytics investment is expected to jump from 4.6% to almost 22% over the next three years, an increase of 376%.

Due to the overwhelming amount of data, it’s easy to get lost and focus on metrics that don’t convert into ROI. Currently, only 22% of marketers say they have data-driven marketing initiatives that are achieving significant results. Therefore, the analytics engine that you choose is immensely important. A segmentation tool like Watson Marketing Insights does the heavy lifting in accurately understanding and segmenting audiences, leaving your marketing team time to take those audiences and create curated experiences.

A good segmentation tool not only groups customers based on past behaviors, it groups them based on predictions about future behaviors. Watson Marketing Insights uses predictive models and cognitive insights to quickly and accurately segment customers based on many customer behaviors, including basic demographic information to transactional and behavioral data – think spending habits, transactional history, returns, customer journey on a website, etc. Including sophisticated data points results in better behavioral targeting, adding to a better customer experience.

Let’s look at an example: Say you’re a marketer for a sportswear company. You want to create a campaign that will build value of your customers, but you don’t know who to target. You have a segment called “Yearly Purchasers” for your customers who have a moderate level of engagement with the brand (they sometimes open emails) and tend to purchase sneakers once a year. You want to appeal to them in ways that will inspire them to run more, and therefore buy more. So you build a campaign with running tips and links to race registration, in addition to running a cross-sell campaign for running tights and jackets. If they catch the running bug, your “Yearly Purchasers” may turn into some of your most valuable customers.

Step 3: Monitor, test, and optimize

We have only cracked the surface on the technological ways to build relationships with customers. Only 10% of US marketers feel as though they have successfully mastered the customer experience. To create a better experience for your customers, you need a continuous feedback loop between your campaigns and analytics tools. When you link them together, results from current campaigns can be fed back into the analytics tools. This creates a feedback loop where campaign performance informs the optimization of audiences. The Watson Marketing portfolio is a complete marketing ecosystem, where marketers can learn and respond in real time as customers habits shift. 


Learn more about how IBM Watson Marketing Insights presents cognitive insights to the marketer, helping to create precise target audiences, tying insights with actions, and driving meaningful business outcomes. See what it can do for your business.

Please note that DISQUS operates this forum. When you sign in to comment, IBM will provide your email, first name and last name to DISQUS. That information, along with your comments, will be governed by DISQUS’ privacy policy. By commenting, you are accepting the IBM commenting guidelines and the DISQUS terms of service.