Study Shows AI Can Optimize Customer Segmentation Using a Powerful Model

Good companies know that customer segmentation is a necessity. It is used for everything from market strategies to discount offers to loyalty program development. The more accurate the segmentation, the more likely the company is to target customers so they come back more often.

For this purpose, many companies turn to RFM, a segmentation model based on three factors: recency, frequency and monetary analysis. Under this system, companies categorize their customers based on how long it has been since they purchased something. how long they wait between purchases; and how much they spend over time.

Now that AI-powered tools and platforms are becoming a staple of businesses, with 72% using AI for at least one function, some are looking to update their RFM strategies with this technology. But with stories still circulating about AI hallucinating or making other errors up to 27% of the time, some business leaders are understandably concerned about trusting it to perform such an important function.

A team of researchers set out to evaluate how well AI can perform at customer segmentation. Malay Sarkar, Faiaz Rahat Chowdhurv and Aisharyja Roy Puja conducted an experiment by making a popular algorithm perform RFM analysis. They then went through the results.

“The experimental results provided convincing evidence of the algorithm’s performance in terms of consumer segmentation,” they wrote in an article published this year in the Journal of Business and Management Studies. They found the “cluster purity score” to be 0.95. Put simply, this meant that the analysis “achieved a relatively high accuracy rate of 95% in terms of precisely and accurately segmenting consumers based on their common behaviors and characteristics… This showed that the algorithm efficiently organized consumers and into different clusters classified their similarities, which enables targeted marketing strategies and personalized approaches.”

Because my work focuses on helping companies deliver the best possible customer experience, I know that many companies are looking to achieve more comprehensive segmentation that includes additional factors. This also proves to be a strength of AI.

In a separate study, six researchers proposed an “extended RFMD model” in which the D stands for “demographic.” In this system, the three financial metrics were combined with demographic information about each consumer, such as age, gender and the region in which they live.

The researchers (Thanh Ho, Suong Nguyen, Huong Nguyen, Ngoc Nguyen, Dac-Sang Man, and Thao-Giang Le) again used algorithms and found that they could divide consumers into further groups, and the results were positive. “Companies can apply this model to accurately understand customer behavior based on their demographics and launch efficient campaigns,” the group wrote in the Business Systems Research Journal.

Beware of the “RFM Trap”

Although these types of analytics are helpful, companies should be careful not to rely on them too much. In a blog post, Data Science Logic – a team of specialists in data analysis and the use of machine learning – warned that companies could fall into an “RFM trap”.

This usually happens for one of two reasons: It is the main or even only tool an organization uses “to analyze and plan communications strategies and activities,” or those tasked with providing insights based on the analysis to win, go too far and reach conclusions that are not really supported by the data.

Companies should always remember that every customer is a unique individual. Predictions and suggestions that AI can provide can have great potential. However, once a customer indicates a preference that does not align with a prediction or expectation, all marketing and sales efforts targeted at that person should change accordingly.

This is another reason why an omnichannel setup is essential. By bringing together everything about a customer’s journey in a single place and using AI to draw instant insights, a company can ensure it treats each customer as an individual. Whether the customer is interacting with a human or a chatbot, this understanding of who they are is proven to increase feelings of empathy and leave people feeling more positive about the company.

AI tools are improving and changing rapidly. Their success with RFM is just the latest sign that this technology is bringing immeasurable benefits and enabling companies to serve their customers better than ever before.