Predicted CLV

Predicted customer lifetime value represents the total value of all orders a customer is predicted to make if they return to make another purchase during the next 365 days.

Amperity models predicted CLV by comparing what customers spent in the previous year to their predicted spend in the coming year, and then determines for each customer their:

  1. Predicted probability of purchase

  2. Predicted number of orders

  3. Predicted average order value

Use predicted CLV modeling to build high-value audiences that identify:

  • Which customers have the highest predicted value?

  • Which customers will respond better to special offers and perks?

  • What are the best personalized experiences for your best customers?

  • Which customers have individual price preferences?

Why not use customer lifetime value (CLV)?

Customer lifetime value (CLV) measures how valuable a customer has been to your company or brand.

CLV is an important historical measurement that applies to individual customers and also to groups of customers that fit various segment profiles. For example, a customer who buys a pair of $40 pants every year for five years has a CLV of $200. A segment that contains thousands of customers who buy pants might have a CLV of $200,000.

CLV can be more complex to measure when a company has millions of customers, multiple brands, storefronts and websites and applications, and hundreds (or even thousands) of individual product items.

How do your brand measure how valuable customers will be in the future? With Amperity your brand can use predictive modeling to identify which customers are more likely (and less likely) to purchase within the next calendar year.

About predicted CLV (pCLV)

Predicted customer lifetime value represents the total value of all orders a customer is predicted to make if they return to make another purchase during the next 365 days.

Predicted customer lifetime value (predicted CLV) sorts customers into the following value tiers:

  • Platinum The top 1% of all customers.

  • Gold 2-5%.

  • Silver 6-10%.

  • Bronze 11-25%.

  • Medium 26-50%.

  • Low The bottom 50% of all customers.

Build segments that capture your most valuable customers, and then build segments that align active customers within those tiers to the right product recommendations and offers.

Building segments for customer lifetime value with Amperity.

Without access to predictive modeling brands are forced to estimate future purchases by using purchases from the previous year. This approach is generally challenging due the presence of one-time buyers, shifting dynamics among customer behavior, changes in inventory and offerings, and competition with other brands.

Building segments for customer lifetime value without Amperity.

Use predicted CLV to identify your top customers by predicted spend, and then avoid having to use only historical purchases when deciding how and when to market to your most valuable customers.

  • Which customers are the most valuable and how can we attract more of them to our brand?

  • Is there anything unique about high value customers?

  • Are dedicated customers for one product more valuable than dedicated customers for another?

  • What can we learn to better inform acquisition efforts and product merchandising?

  • Does a promotion attract high or low value customers?

  • Should a promotion be relaunched next month, next quarter, next year?

  • Which customers are best engaged through a VIP program?

Models for pCLV

Predictive modeling for customer lifetime value is comprised of several component predictions:

Predicted probability of transaction

Predicted probability of transaction represents the likelihood that a customer will return to make another purchase during the next 365 days.

For example, a score of 0.6 indicates a 60% chance that a customer will return and make purchase.

Note

Predicted probability of transaction is also known as p(return).

Predicted average order revenue

Predicted average order revenue represents the average value of each order a customer is predicted to make if they return to make another purchase during the next 365 days.

Predicted order frequency

Predicted order frequency represents the number of orders a customer is predicted to make if they return to make another purchase during the next 365 days.

Start using predicted CLV

Start with a combination of transaction attributes, apply predicted CLV, and then group them by customer lifecycle status, typically using active or cooling down, but sometimes using at risk. The following segments represent different types of customers who are likely to respond to a message in different ways:

  1. A segment for customers who have purchased more than once and for whom order revenue in the previous year is greater than $500 and with an active lifecycle status.

  2. A segment for customers who have purchased just once in the previous year with an order value less than $200 and an at risk lifecycle status.

Talk with your Amperity representative about the best attributes to use within your tenant to enable effective campaigns that use predicted CLV.

Use cases

The predicted CLV model helps you identify your highest value customers by year or by value tier:

  1. How much will customers spend in the next year?

  2. Which customers are your most valuable customers?

How much will customers spend?

The Predicted CLV Next 365d attribute in the Predicted CLV Attributes table contains the total predicted customer spend over the next 365 days. You can access this attribute directly from the segment editor:

Choose the predicted CLV attribute from the Segment Editor.

After you select this attribute you can specify the type of values you want to use for this audience, such as:

  • Predicted value is greater than $100

  • Predicted value is less than $400

  • Predicted value is between $100 and $400

Which customers are the most valuable?

When predictive modeling is enabled for your tenant you can use output from the predicted customer lifetime value (CLV) model, which helps you identify your highest value customers by value tier. Each tier represents a percentile grouping of customers by predicted value:

  • Platinum represents the top 1%

  • Gold represents customers who fall between 1% and 5%

  • Silver represents customers who fall between 5% and 10%

Select all three of these predicted value tiers to build an audience that contains customers who are predicted to be in your top 10% (inclusive) high value customers.

The following sections describe using the Segment Editor to build a segment that returns customers who are predicted to be your top 10% highest value customers.

WHICH CUSTOMERS ARE PREDICTED TO BE YOUR TOP 10%?

Open the Segment Editor.

Open the Segment Editor, look in the lower-right of the page and make sure your customer 360 database is selected.

Use your customer 360 database to build segments.
Return a list of the customers with a predicted platinum, gold, or silver value.

The first step is to identify customers whose predicted customer lifetime value is platinum, gold, or silver. Choose the Predicted Customer Lifetime Value Tier attribute from the Predicted CLV Attributes table, select the “is in list” operator, and then select “Platinum”, “Gold”, and “Silver” from the list:

Find customers with a predicted platinum, gold, or silver value.

Click the Refresh button located on the right side of the Segment Editor to see how many customers are in your segment, how much they spent in the past year, how many are active, and how many of them should belong to a future campaign.
Return a list of customers for whom your brand has email addresses.

The next step is to identify customers with contactable email address in their customer profiles. Choose the Contactable Email attribute from the Customer Attributes table, and then select the “is true” operator:

Find customers for whom your brand has email addresses.

Keep the and/or slider set to AND.

Click the Refresh button located on the right side of the Segment Editor to view updated values for the combination of customers who have a predicted platinum, gold, or silver and a contactable email address.

Tip

Use the Is Opted Into Email attribute from the Customer Attributes table to include only customers who are opted into receiving email messages from your brand.

Find customers for whom your brand has an opted-in email address.
Send customer list to your favorite email marketing tool.

Send this list of customers to your favorite email marketing tool (i.e. Attentive) on the Campaigns page.
Save your segment.

You’re done building your audience! Click the Save As button in the top right corner of the Segment Editor. Give your segment a name that clearly describes the purpose and audience type for the segment. For example: “Predicted Top 10% High Value Customers”

Give your segment a name.

Tip

Use good naming patterns to ensure that you can always find your segments when you need them. Be sure to include the brand name and/or the region name if you have multiple brands or have multiple regions and want to build segments that are brand- and/or region-specific.
Segment insights page

After your segment is saved the Segment Overview page opens and shows additional details, such as historical and predicted revenue, the percentage of customers that are reachable by email, by phone, on Facebook, and customer trends, such as purchases by channel, revenue by lifetime spend.