Can AI-driven segmented marketing generate increased revenue? Let’s find out!

This is the first article in a series of customer case studies and use cases that I will be writing over a period of coming weeks/months.

My first post focuses on using AI to create targeted customer segments. Next post will cover the application of embedded machine learning to prevent customer churn. I’ll be using a sample customer case study to define the need, problem, solution, approach and key benefits. We will also go over customer’s improved business outcome realized by leveraging the Tellius platform.

Let’s start with the first case study — Segmented Marketing to improve Revenue.

What is customer segmentation?

Customer segmentation is the practice of grouping customers. Typically, it is done in such a way that individuals in a specific group depict similar characteristics such as age, gender, interests and spending habits. These segments are then put to use in various ways such as targeted marketing.

The system of belief is that customer segmentation can help reach specific groups of individuals (small in size) with very personalized messaging. The messaging can then be fine-tuned such that the individuals in these groups would find the messaging content very relevant to them. The theory is that such hyper-personalized targeting will lead the individuals into buying something that you are selling.

There is no doubt that it definitely makes sense to spend marketing efforts (and $$$) on targeted messaging v/s doing a blast message with low ROI. In fact, one of the research concluded with this data:

“Segmented marketing campaigns generate an average of 760% increase in revenue¹”

The concept sounds reasonable and looks simple to implement — but only in theory. The real challenge marketing teams run into is the creation of such targeted segments.

Marketing teams don’t have any easy-to-use tools at their disposal to help identify key differentiators that divide customers into groups to be targeted. Most of it today is based on tribal knowledge / biased judgment. But such processes can only take you so far — especially with the ever-changing customer demographics, spending patterns, new product introduction etc.

So, that brings us to the meat of discussion — now that we know that there’s high ROI on targeting customers via segmentation, how do we get there? Let’s talk about a very specific problem one of our customers came to us with recently.

Can AI-driven Analytics help Marketers to optimize campaigns?

So we will dive deeper into a case study that lays out the need, problems, solution, approach and benefits for customer segmentation. Data, names, numbers and a few other items have been anonymized/randomized to protect confidential information and respect their privacy.

Business: Top supermarket chain.

Industry: Retail.

Primarily, selling consumer goods (and/or services) to customers through multiple channels of distribution to earn a profit is their core business.

Supermarket chains are usually among the largest retailers in the world. The market is cannibalized by a few billion-dollar multinational companies. Big retailers have set up huge supply/distribution chains, inventory management systems, financing pacts and wide-scale marketing plans.

Need: As we can see, the customer’s marketing team operates in a highly competitive market. They need to reach out customers who are at risk of leaving. They want to increase the wallet share. They want to upsell to their customers. And so on…

Problem: Since last year, the client started to gather customer data through a loyalty program, that tracks each customer’s purchase. Campaigns were prepared on a daily basis. Advertisement magazines and emails were sent to every customer in each campaign.

Turns out, the supermarket was spending way too many resources into contacting customers that weren’t interested in specific campaigns. The reach out started annoying their customers as the content was generic/irrelevant and sent frequently.

So their real problem was to personalize campaigns and target specific groups. They wanted help in identifying precise customer groups for each campaign.

Solution: With Tellius, supermarket’s marketing analysts started discovering personalized customer segments using the underlying sophisticated machine learning algorithms. Customers were clustered according to their behavior and socio-demographic patterns. The marketing department now utilizes these clusters for new campaigns and therefore only contacts customers that show interest in the subject.

Approach: Here’s the three-step approach that worked really well:

  • Identify and connect to the right set of data: Loyalty and Purchase. Key information in these datasets: Social, Demographics, Behavior, Purchase, Recency, Frequency, Monetary Spend, Product Purchased, etc.
  • Self Service data transformation: Unimportant columns were removed, Missing/Invalid data fixed using quick options, Data from various sources joined together.
  • Single Click Automated Insights: Insights were discovered that generated targeted clusters of customers. Predictive characteristics of various segments and their propensity to respond/buy various products were predicted.

Key findings discovered by using Tellius platform

Some interesting insights were automatically surfaced by the Tellius Genius AI engine.

These discoveries amazed the supermarket’s marketing team. I’ll list a few of the interesting ones below:

  1. Customers spending above average on Toys have a high propensity to spend on / buy magazines, housing tools, food, cleaning items, drinks and baby products. It can be seen in the Top Predictive Factors card in the image above.
  2. Customers with a 4x times higher chance of buying toys were discovered. Similar customer segments were created for every product group. It can be seen in the Top Recommendation card in the image above.
  3. Influential factors were discovered for each product category. Precise relevant segments that focus on profitable customers for each of these product groups were auto-generated.
  4. Some interesting trends on spending patterns of different age groups across various product categories were discovered. In general, there was a linear growth between the average amount of money spent and the age of a customer. But, some product categrories were outliers where the trend had a few baseline changes as age progressed. As an example, this was automatically highlighted in the Toys v/s Age spending Insight.


As we saw in the case study above, one of the top supermarket chains was able to utilize Tellius’ platform to optimize their campaign performance and improve the ROI. Note that all of the insights were discovered automatically without spending 100’s of hours on manual data discovery or without wiriting a single line of code.

Tellius platform utilized it’s underlying machine learning algorithms to answer specific business questions and discover hidden insights in the data.

The insights discovered and algorithms developed won’t be thrown away. The algorithm actually learns from the past customer patterns and applies this knowledge automatically to the current and future customers. The machine learning model will be trained at regular intervals as and when new customer data is made available. This will improve the overall performance of the customer segmentation predictive model.

Now, the marketing department exactly knows which advertisement campaigns to run to have a specific influence on various product groups. They’ve started saving on resources by running very targeted and personalized campaigns. Only customers in the target segment for each product group are contacted, which increases the ROI of each campaign enormously.

The supermarket chain can now:

  • Adjust its overall strategy to become more attractive to specific customer groups.
  • Reduce campaign cost as less number of customers will be contacted for marketing campaigns.
  • Improve profits as only interested customers will be targeted by the campaigns.
  • Achieve higher ROI — time and cost/savings, and increase in revenue.

— -

Hope this post was useful to understand how advances in machine learning can be applied to solve real-world problems. Shortly, I’ll start working on the next article, until then — stay tuned!

— -

Please feel free to reach out directly for feedback, questions, and comments:

— -

Note: Any references or images used in the post are only for informational purposes. The ownership and copyright remain with the original creator. Please let me know if any content violates your rights — I will take it down as soon as possible.