SoundCommerce Customer Lifetime Value (CLV)

A recent addition to the SoundCommerce Customer data model, Customer Lifetime Value (CLV) is a key determinant of a company’s longterm viability. CLV represents the profit value a customer generates during their lifetime. Lifetime is defined as the time between first purchase and churn.

By understanding CLV, you can drastically increase the value of your existing customer base and acquire customers who have the potential to become high-value, repeat customers. Marketers frequently use CLV to make informed decisions around ad spend and keep customer acquisition costs (CAC) in check.

In this article:

Data requirements

The CLV data model and dashboards are included in the following module combinations:

SoundCommerce Campaign with SoundCommerce Profit
SoundCommerce Customer with SoundCommerce Profit

How we calculate CLV

The SoundCommerce CLV data model defines CLV for a unique customer (or CUID) as the sum of historic profit and predicted future profit. If cost data is unavailable, the alternative to CLV is Customer Lifetime Sales (CLS).

Customer Historic Profit is the sum of contribution profit across all previous orders. For our predictive model, we use the past three years of historic data.

Customer Predicted Profit is the sum of contribution profit across orders the customer is expected to make in the next 12 months. Customer predicted profit is a dynamic metric that is calculated daily for every customer. For example, if the current date is May 12, 2022, Customer Predicted Profit will be the sum of expected profit through May 11, 2023.

To get the contribution profit for a single order, SoundCommerce accounts for all available variable costs (COGS, shipping costs, and ad spend). Variable cost data often comes from multiple ERP, carrier, and logistics systems—all of which require integrations. As you provide more data to SoundCommerce, the accuracy of contribution profit improves.

Direct Access Fields

Data Kind	Direct Access Field	Data Type	Description
Customer	CLV	Numeric	Customer historic contribution profit plus predicted contribution profit for the next 12 months
Customer	CLV_segment	String	Low, Medium, or High (Output of k-means clustering using single dimension of CLV)
Customer	predicted_profit	Numeric	Predicted contribution for next 12 months updated daily
Customer	predicted_profit_segment	String	Low, Medium, or High (Output of k-means clustering using single dimension of predicted_profit)
Customer	CLV_CAC	Numeric	CLV / CAC
Customer	CLV_accuracy	Numeric	Accuracy of the CLV segment assigned to the customer; segment accuracy is defined as number of correct predictions over total predictions
Customer	profit_accuracy	Numeric	Accuracy of the predicted profit segment assigned to the customer; segment accuracy is defined as number of correct predictions over total predictions

Predictive profit metrics

Predicted profit is calculated for each customer record (CUID) using a machine-learning regression model. By analyzing three years of historical order and cost data, the CLV model can predict contribution profit for the next 12 months.

Profit predictions associated with any predictive model are most accurate when applied as an average across several customers. Applied to individual customers, they sometimes miss the mark by over or under predicting profit.

By measuring the average CLV and average predicted profits across all CUIDs in a segment, you can dramatically improve the accuracy of your profit predictions. For example, if our predictive metrics tell us that cuid_1119 has a CLV of $50, the customer's actual CLV may end up above or below this threshold. While we can't predict the CLV for cuid_1119 with absolute certainty, we can analyze several customers who are similar to cuid_1119, and reliably predict their collective CLV to be $40-60.

Quicker results with segments

To provide quicker insights as to where customers fall on the spectrum of predicted profit and lifetime value, we've provided six pre-built customer segments. Each segment uses k-means clustering to aggregate customers into groups of Low, Medium, or High, for both CLV and predicted profit, based on their individual data. We explain this process in How segments are built with k-means clustering.

From the CLV dashboard, you can see how many low CLV customers you have compared to high CLV customers and how segments are defined in terms of contribution profit. From there, you can take immediate action in Query Builder to investigate causes of low CLV and target these customers through email or SMS marketing campaigns.

Whether you use Query Builder or Direct Access, pre-built segments let you build customer lists based on more accurate predictions. To view these segments, go to the CLV dashboard.

The CLV dashboard shows you accuracy ratings, the mean CLV, and the range of values for each customer segment.

How accurate are CLV segments?

Each pre-built segment includes an accuracy metric that tells you, of the customers predicted to fall in a segment, how many actually land in that segment. Therefore, accuracy is defined as the percentage of correct predictions divided by the total number of predictions.

For an example of accuracy metrics, we can look at the Low predicted profit segment for our fictional retailer, Jet City Organics. On April 14, 2022, the accuracy of the Low predicted profit segment was 94%. This tells us that 94 out of every 100 customers who were predicted to have Low contribution profit a year ago actually fell in the low range.

Directly below the graph is the accuracy rating for each segment (Low, Medium, High). Order profitability is typically more accurate for segments versus individual customer IDs.

Keep in mind that the accuracy metric will fluctuate. This is because the dataset we use to generate predictions updates on a daily basis.

Search