Customer Lifetime Value
Introduction
Detecting early signs of reduced customer engagement is pivotal for businesses aiming to maintain loyalty. A notable signal of this disengagement is when a customer's once regular purchasing pattern starts to taper off, leading to a significant decrease in activity. Early detection of such trends allows marketing teams to take swift, proactive measures. By deploying effective retention strategies, such as offering tailored promotions or engaging in personalized communication, businesses can reinvigorate customer interest and mitigate the risk of losing them to competitors.
Our objective is to utilize a model that not only alerts us to customers with an increased likelihood of churn but also forecasts their potential purchasing activity and, importantly, estimates the total value they are likely to bring to the business over time.
These analytical needs are served by what is known in data science as Buy 'Til You Die (BTYD) models. These models track the lifecycle of a customer's interaction with a business, from the initial purchase to the last.
While customer churn models are well-established within contractual business settings, where customers are bound by the terms of service agreements, and churn risk can be anticipated as contracts draw to a close, non-contractual environments present a different challenge. In such settings, there are no defined end points to signal churn risk, making traditional classification models insufficient.
To address this complexity, our model adopts a probabilistic approach to customer behavior analysis, which does not rely on fixed contract terms but on behavioral patterns and statistical assumptions. By doing so, we can discern the likelihood of future transactions for every customer, providing a comprehensive and predictive understanding of customer engagement and value.
Customer Lifetime Value - How it Works?
The Customer Lifetime Value (CLV) model is a robust tool employed to ascertain the projected revenue a customer will contribute over their entire relationship with a business. The model employs historical data to inform predictive assessments, offering valuable foresight for strategic decision-making. This insight assists companies in prioritizing resources and tailoring customer engagement strategies to maximize long-term profitability.
The CLV model executes a series of sophisticated calculations. Yet, its operations can be conceptualized in a straightforward manner:
Historical Analysis: The model comprehensively evaluates past customer transaction data, noting the frequency and monetary value of purchases alongside the tenure of the customer relationship.
Engagement Probability: It assesses the likelihood of a customer’s future engagement based on their past activities, effectively estimating the chances of a customer continuing to transact with the business.
Forecasting: With the accumulated data, the model projects the customer’s future transaction behavior, predicting how often they will make purchases and the potential value of these purchases.
Lifetime Value Calculation: Integrating these elements, the model calculates an aggregate figure representing the total expected revenue from a customer for a designated future period.
Model Scenario
To set up the model, you’ll need to configure a few key fields:
Time/Date Column: Select the column that contains the date of each transaction (e.g., invoice date, order date). This tells the model when each customer activity occurred.
Customer ID: Choose a column that uniquely identifies each customer. This ensures that all purchases are correctly grouped under the same customer.
Customer Name (optional): If your dataset includes customer names, you can select this field to display names in the results. If not, simply use the same column as Customer ID.
Monetary (amount spent): Select a numeric column that shows the amount spent per transaction. This is the main metric used to calculate lifetime value.
Starting Date for CLV Calculation: Decide from which point in time you’d like to start calculating the customer lifetime value. You can choose:
Max Date: Uses the latest date available in your dataset.
Today: Starts the CLV calculation from the current date.
Custom Date: Manually select any specific date to serve as the calculation start point.
Model Results
Once you’ve run your CLV scenario, the Results tab will present a comprehensive view of how valuable your repeat customers are, how long they are expected to remain active, and what you can expect in future revenue. The Results section is split into four core views:
Each view gives you unique perspectives on your customer base and revenue predictions. Below is a guide for interpreting each one.
Overview
The Overview tab gives you a high-level summary of your repeat customers and their expected future behavior, all in one place.
At the top, you’ll see key metrics that describe the health and value of your customer base:
Total Repeat Customers – how many customers made more than one purchase
Total Historical Amount – the total revenue generated by these repeat customers
Average Spend per Customer – average revenue per repeat customer
Average No. of Repeat Purchases – how many purchases each customer made on average
Average Probability Alive (Next 90 Days) – how likely they are to stay active in the near future
Predicted No. of Purchases and Amount (Next 90 Days) – expected orders and revenue in the next three months
CLV – estimated lifetime value per customer
These numbers give you a snapshot of customer loyalty, revenue contribution, and future potential. Below that, in Deep Dive into Customer Behavior section, you’ll find a breakdown of how these insights are calculated. It includes values like total revenue per customer, number of purchases, average spend, customer age, and their predicted likelihood of staying active.
Overview helps you better understand which customers are most engaged, most valuable, or at risk, so you can take action where it matters most.
CLV Insights
This tab offers time-based forecasts in visual form to help you understand how customer behavior and value change over time. Two main charts are presented:
Forecasted Number of Purchases
A line chart showing how many purchases are expected in 7, 30, 60, 90, and 365 days.
The curve reflects expected buying activity, helping you estimate short- and long-term customer engagement. A steeper curve suggests strong future buying behavior, while a flatter line indicates less projected activity.
Forecasted Amount
This chart mirrors the number of purchases, but focuses on revenue. It shows how much your customers are likely to spend within the same time intervals:
7, 30, 60, 90, and 365 days.
This helps you plan revenue forecasts and guide sales or marketing campaigns based on expected spend. For example, a sharp increase toward 365 days may indicate strong long-term potential.
The average alive probabilty
This chart illustrates the average probability that a customer remains active (i.e., still engaged and likely to purchase again) over time — assuming no repeat purchases.
The curve shows how this probability gradually declines as time passes without interaction. For example, if the probability drops significantly by 365 days, it signals a potential churn issue in the long run.
Together, these insights give you a forward-looking view of how much value your existing customers will bring in both volume and revenue, so you can plan smarter.
Segments
The Segments tab provides a visual breakdown of your customer base according to their likelihood of remaining active.
The Customer Segmentation Based on Probability Alive pie chart groups customers into five risk levels—Very Low Risk, Low Risk, Medium Risk, Very High Risk, and High Risk—based on their probability of staying active. This segmentation helps you easily identify which customers are most likely to churn and which are loyal.
The Customer Churn Risk Analysis bar chart simplifies this further by splitting customers into two groups using a 50% probability threshold: those more likely to stay and those more likely to leave. This chart offers a quick snapshot of retention risk across your entire repeat customer base.
Together, these visuals help you prioritize your marketing and retention efforts by highlighting which segments need attention and which are driving your recurring revenue.
Details
The Details tab presents a full breakdown of all customers used in the Customer Lifetime Value (CLV) model, showing individualized metrics for each one. This table allows you to explore how the model calculates and segments customer lifetime behavior on a granular level.
Each row represents one customer, and each column offers specific insight into their engagement, loyalty, spending behavior, and predicted future value. You can filter or search by any column to find customers of interest (e.g., high spenders, churn risks, new users, etc.).
Key Columns Explained:
Here are some of the most relevant fields available in the table:
amount_sum - Total amount spent historically by the customer.
amount_count – Number of total purchases made.
repeated_frequency – How many of those purchases were repeat purchases.
customer_age – Age of the customer in days since their first transaction.
average_monetary – Average amount spent per transaction.
probability_alive – Current probability that the customer is still active (1 = 100%).
probability_alive_segment – Segment grouping based on risk (e.g., Very Low Risk, High Risk).
probability_alive_50_perc – Whether the customer’s probability to stay is above or below 50%.
predicted_no_purchases_7_30_60_90_365 – Forecasted number of purchases the customer is likely to make in the next 7, 30, 60, 90, or 365 days.
CLV_30_60_90_365 – Estimated monetary value the customer is likely to bring in the next 30, 60, 90, or 365 days.
avg_days_between_purchases – Average time between each purchase.
days_since_last_purchase – Time since their last transaction.
Each metric helps you evaluate customer loyalty, frequency, predicted churn, and future potential revenue. You can export this data into Excel for further analysis or integrate it with marketing and CRM workflows. Because each row represents one individual customer, and all the key metrics are already calculated (such as purchase frequency, churn risk, and expected revenue), you can immediately use this information to guide your next steps.
Using Details to Understand Individual Customer Behavior
The Customer Lifetime Value (CLV) model helps you see beyond averages by offering detailed insights into each customer’s unique behavior. Let’s compare two very different shopping profiles found in model results.
Customer 1 – Loyal and Consistent
This customer shows clear signs of long-term engagement:
They’ve been active for nearly 4 years (1,474 days),
Have made 7 purchases, of which 6 are repeat transactions,
And contributed over €1,100 in total revenue.
What stands out most is their very high probability of continued activity—over 98%, both now and over the next 7 days. The predicted Customer Lifetime Value (CLV) over the next year is also significant. This customer is a clear example of high loyalty, high predictability, and strong recurring value.
Customer 2 – Occasional and Unpredictable
By contrast, this customer represents a much riskier profile:
Despite being in the system for 3 years (1,097 days), they’ve only made 2 purchases,
With just 1 repeat,
And a total spend of around €380.
Even though their average transaction value is quite high, their behavior is infrequent and highly unpredictable. The model assigns them a low probability of being active (33%), both currently and in the upcoming week. Their 1-year CLV forecast is modest, and they fall into a high-risk segment.
What This Tells Us?
This comparison highlights how businesses can tailor engagement strategies:
For Customer 1, the goal is retention and appreciation—they’ve proven loyal and should be nurtured with rewards or exclusive offers to maintain momentum.
For Customer 2, the opportunity lies in re-engagement—offering incentives or personalized outreach could help turn a sporadic buyer into a more regular one.
By understanding these behavioral patterns, teams can move beyond surface metrics and take targeted, meaningful actions to increase long-term customer value.
Take actions with CLV
The true power of Customer Lifetime Value (CLV) modeling lies not only in understanding historical customer behavior but in using that understanding to shape future strategy. Just as binary classification allows us to distinguish between likely churners and loyal customers, CLV reveals how much value each individual customer is expected to generate—and how likely they are to remain engaged.Here’s how to utilize the information effectively:
Custom Segmentation: Use
customer_age,amount_sum, andaverage_monetaryto segment your customers into meaningful groups.Detect Churners: Use
probability_aliveto segment customers currently being active for non contractual business like eCommerce and Retail. A score of 0.1 means 10% probability the customer is active ("alive") for your business.Targeted Marketing Campaigns: Leverage
repeated_frequencyandprobability_alivecolumns to identify customers for loyalty programs or re-engagement campaigns.Revenue Projections: The
CVL_30_60_90_365column helps in projecting future revenue and understanding the long-term value of customer segments.Strategic Planning: Use
predicted_no_purchases_7_30_60_90_365to plan for demand, stock management, and to set realistic sales targets.
By engaging with the columns in the Details Tab, users can extract actionable insights that can drive strategies aimed at optimizing customer lifetime value. Each metric can serve as a building block for a more nuanced, data-driven approach to customer relationship management.
Actionable Insights
The Actionable Insights section translates complex CLV model results into clear, strategic takeaways. It identifies which customer segments pose the greatest revenue risk due to low retention probabilities and quantifies the financial impact of improving retention. By simulating small increases in customer engagement across segments, it reveals the potential uplift in revenue and highlights the most effective levers—such as win-back offers, onboarding nudges, or tailored campaigns. This enables businesses to prioritize actions where they’ll generate the most value, turning predictive insights into measurable impact.
Model Results API
All CLV model results can be retrieved programmatically using the Model Results API. This is especially helpful when:
You want to analyze CLV outputs or take actions in external tools (BI or CRM)
You need to trigger automated actions based on CLV scores—for example, enrolling a “Very High Risk” customer in a retention campaign.
The API returns structured outputs for each customer, including total spend, purchase frequency, risk segment, and predicted future value—enabling custom workflows, real-time applications, or dashboard integrations.
For implementation details, refer to the Model Results API documentation.
Notebooks
With the Notebook feature in Graphite Note, you can present CLV model outputs in a compelling, shareable format. These notebooks are ideal for team collaboration, reporting, or embedding into partner-facing portals. Notebooks act as your storytelling layer—bringing data to life and turning predictions into strategy. For more information, refer to the Data Storytelling section.
Last updated
Was this helpful?