Powering Strategic Decisions with Data: Senior Leaders’ Roadmap to CRM Analytics
CRM, RFM, CLV… These terms might not be unfamiliar to you, and you might even be crafting strategies using these data points. Fantastic! Maybe these concepts aren’t entirely new to you, but you’re diving into customer segmentation and strategy formation based on it. Or perhaps this is entirely new territory, and you’re keen on starting from scratch…
Regardless, in this piece, I’ll break down CRM analytics for you simply and understandably. By the end, you’ll uncover how to harness customer data effectively. Let’s dive right in!
CRM — Customer Relationship Management
No matter the size of your business, if you’re an owner making sales, the first thing you need to tackle is Customer Relationship Management. It’s a general concept that expresses the art of handling customers. A simple term encompassing the customer management of companies, be they with products or services, and their sales processes.
This covers all processes related to customers:
· Customer lifecycle optimizations: This involves the strategies or funnels that need to be tracked from the first contact with the customer to the first purchase. UX and UI are crucial in this step.
· Communication: The tone when speaking with customers, or the tonality applied in written forms. The choice of colors and visuals on your website or banners, and so on.
· Customer acquisition efforts.
· Customer retention or churn campaigns.
· Cross-selling, upselling.
· Segmentation studies.
Based on these steps, you create various groups/segments/personas. For example, you can form segments like “buyers,” “new members,” or “non-purchasers.” Then, you focus on these segments separately, crafting strategies to increase sales, prevent customer churns, or win back customers.
In essence, if a company has products and is involved in sales, the first place to analyze should be its own CRM.
CRM Analytics
CRM Analytics is the data-driven approach to handling the customer communication processes I’ve listed above in bullet points.
Virtually all companies touching on CRM analytics kick off with the following three core practices:
1- Customer segmentation with RFM (Recency, Frequency, Monetary)
2- Customer Lifetime Value (CLV or CLTV)
3- Customer Lifetime Value Prediction
The fundamental drive behind CRM Analytics:
• To segment customers,
• To devise distinct strategies for these segments,
• To make projections for campaigns, sales, or budgeting efforts.
Or with Vahit Keskin’s one-sentence explanation:
CRM Analytics helps us make decisions among uncertainties with the least error.
Let’s delve into how we can achieve these motivations.
1- Customer Segmentation with RFM
RFM is a rule-based method for customer segmentation. This concept is represented by the initials of recency, frequency, and monetary
- Recency — How recently did the customer purchase?
- Frequency — How often does the customer purchase?
- Monetary — How much does the customer spend?
Let’s say your boss says, “Prepare an action plan (or campaign) for our highest-paying customers.” What would you do? To find the highest payers, you would examine the purchase amounts for each customer, rank them from highest to lowest, and identify who has contributed the most. Now, how many of these high-payers are there? How many individuals are you planning to take action for? To determine this, you might consider the following:
1- Progress with a ratio. You could aim for the top 3% or 5% of the highest payers.
2- You might use the Pareto principle (the rule 80/20) or,
3- You can segment all your customers by conducting RFM analysis.
Consider having a sales dataset like the one below and let’s segment them using RFM analysis:
Step 1: Calculating Recency Score
• Determine the most recent purchase date for each customer.
• Subtract the most recent purchase date from a specific date you want to analyze. This will give you the “Recency” value for each customer.
• Assign a score of 5 to customers who made the most recent purchases and a score of 1 to those who made purchases long ago. For instance, customers who made a purchase one day before the analysis date would receive a score of 5 (max score), while customers who made a purchase 100 days ago would receive a score of 1 (min score).
Step 2: Calculating Frequency Score
• Determine the total number of purchases for each customer.
• Give a score of 1 to customers with the lowest purchase frequency and a score of 5 to those with the highest purchase frequency.
Step 3: Calculating Monetary Score
• Determine the total purchase amount for each customer.
• Assign a score of 1 to customers with the lowest spending and a score of 5 to those with the highest spending.
For all these steps, you can use the qcut function.
But what is qcut function? It’s a function from the Pandas Library that divides a dataset into groups of equal size. Imagine you have a dataset of 100 elements and you want to divide it into 5 groups of equal size. The qcut function would split this dataset into 5 groups, each containing 20 elements. This ensures that each group has the same size. This way, we can continue our operations, maintaining a scale where the highest score is 5 and the lowest score is 1.
Step 4: Calculating RFM Score
• Concatenate each customer’s Recency, Frequency, and Monetary scores side by side to obtain the RFM score. For instance, if a customer has a Recency score of 5, a Frequency score of 3, and a Monetary score of 4, their RFM score would be 534.
RFM scores are ready! Now it’s time for segmentation. Let’s segment your customers by placing their recency and frequency scores on separate axes.
You might wonder why we didn’t consider the monetary value when creating the table above. The reason is that we are interested in the “frequent” customers — those who keep coming back for more. A customer who frequently purchases from us is more valuable than someone who makes a one-time large purchase and never returns. Think of that customer as a tourist passing by. This is why we use the two-dimensional aspect of the RFM axes.
We’re also adding the segment names from the RF table to our sales data, and the final view of our data takes the following shape:
Coming back to our initial question, our most profitable customers are those in the “Champions” segment. Within the analyzed period, they’ve made purchases both frequently and recently.
— Now, the second question: which other segments in the RF table should we focus on?
• The segments with the highest value of Frequency.
— But why?
• Customers in these segments (“Can’t Lose Them”, “Loyal Customer”, and “Champions”) have made purchases from us very often. It’s well-known that acquiring new customers is more costly than retaining the ones we already have.
On the other hand, even though “Can’t Lose Them” customers haven’t purchased from us recently, they used to do so frequently in the past. In this case, we should aim to win these customers back and develop new strategies to move them into the “Loyal Customer” segment.
Similarly, specific strategies can be devised for each segment in the table.
2- Customer Lifetime Value (CLV or CLTV)
Customer Lifetime Value is the product of the average purchase amount and the total number of purchases for each of your customers.
So, what benefits does knowing the CLV (or CLTV) value of each customer bring us?
1- Strategic Decision Making: CLV is used to support strategic decisions of our business. Companies use CLV data to devise strategies that better reach their business objectives.
2- Increasing Customer Loyalty: By offering special deals and promotions to customers with high CLV, you can increase customer loyalty. Loyal customers shop more frequently and tend to remain customers for a longer period.
3- Efficient Marketing Expenses: Focusing on customers with high CLV allows you to use marketing expenses more efficiently.
4- Product and Service Development: Analyzing high CLV customers can help determine which of your products or services sell better and what types of products or services our customers prefer. This information is crucial for improving products and services to meet customer demand.
5- Measuring Financial Performance: You can use CLV to understand the value of our customer database and evaluate our financial performance. Having a high number of high CLV customers indicates long-term success while having a high number of low CLV customers signals a need to review customer retention strategies.
6- Reducing Customer Churn Rate: Developing strategies to transform low CLV customers into more valuable ones helps reduce your customer churn rate.
In summary, Customer Lifetime Value helps you embrace a customer-centric approach and is a tool that assists you in achieving your KPIs.
3- Customer Lifetime Value Prediction
Customer Lifetime Value Prediction is the estimation of the total expected spending over a customer’s long-term relationship with a business.
We use the BG-NBD (Beta Geometric/Negative Binomial Distribution) and Gamma-Gamma models for Customer Lifetime Value (CLTV) prediction because they offer a comprehensive approach to understanding and forecasting customer behavior and value. Here’s why these models are valuable for CLTV prediction:
BG-NBD (Beta Geometric/Negative Binomial Distribution):
- BG-NBD is used to predict customer loyalty and purchase frequency.
- It helps forecast how often new customers will make purchases and how long they will remain loyal.
- This model combines “geometric distribution” and “negative binomial distribution.”
Gamma-Gamma:
- The Gamma-Gamma model is used to estimate the average purchase value of customers.
- It predicts the average spending value based on a customer’s purchase frequency and average spending.
- When combined with the BG-NBD model, it enables CLTV prediction.
By utilizing these two models together, it becomes possible to forecast the future value of customers based on both their purchase frequency and spending behavior. These predictions assist businesses in developing personalized strategies for customers and optimizing their marketing efforts. For instance:
1- Personalized Marketing Strategies: It can be used to create personalized marketing strategies for your customers.
2- Reducing Churn Rate: By predicting how long your customers might stay connected to your business, you can develop strategies to reduce the churn rate.
3- Return on Investment (ROI) Analysis: It helps you evaluate the return on investment for your marketing and customer relationship efforts.
Summing Up the Whole Article:
— The initial data a company often possesses is its sales data. One of the first analyses that can be applied to this data is CRM analytics. The fundamental goal here is:
- Determining how much each customer has paid,
- Can we create segments/personas? RFM analysis is one of these methods.
— What is CLTV and why do we need such a concept:
- CLTV is the revenue a customer will generate for us.
- It’s essential for taking action or positions. While assessing our company’s revenue could provide some positioning, if we want to take actions/positions specifically tailored to individual customers, we can segment individuals and conduct campaigns based on these segments. So, it holds crucial significance for delving into individual specifics.
— CLTV Prediction, on the other hand, is used to calculate the return on strategies we develop for customer-focused actions. If we aren’t achieving the anticipated revenues, it signals the need to refine our strategy.
In the ever-evolving landscape of business, where uncertainties often cloud decision-making, the power of data-driven insights cannot be underestimated. CRM Analytics emerges as a guiding light, illuminating the path for senior leaders seeking clarity in strategy crafting. By unraveling the complexities of customer relationship management, understanding the nuances of customer segments, and harnessing the predictive potential of CLTV, senior leaders are equipped to not only make informed decisions but also steer their organizations toward growth and success. With CRM Analytics as a compass, the journey through uncertainties transforms into a strategic voyage towards excellence. So, let’s embrace the wealth of data at our fingertips and set sail on the transformative journey that CRM Analytics promises.
