BY BROOKE NIEMIEC – CHIEF MARKETING OFFICER
A while back, my colleague Chuck posted an article on how segmentation, in its many different forms, is everywhere. It’s everywhere because it’s useful—a form of classification that helps us categorize things in a meaningful way. It’s also something we all use in our personal lives.
Take a moment and think about your social network. My guess is that you probably know which friends you would invite to a movie, which ones you would take on a white-water rafting adventure, and which ones you would send a birthday gift to. I would also guess that those lists do not contain completely distinct groups of individuals—some of your friends might be members of several of those lists, and others may only fit into one.
Customer segmentation is no different. There are countless ways to create groups of customers based on similar characteristics. One “master” segmentation won’t cut it anymore—different types of segmentation are required to answer different business questions. But where do you start?
Below are six common segmentation types and a brief definition of each.
Behavioral: Tied to actual customer behavior, based on variables such as what customers buy, how frequently they visit, and how much they spend.
Attitudinal: Based on self-reported beliefs and attitudes, often about the brand, category, or lifestyle.
Lifecycle: Used for managing the entire customer journey with the brand over time, from acquisition to retention to reactivation.
Geo-demographics: Broadly applicable segments that capture customer data by geographic region and/or by socioeconomic factors such as gender, income level, or age group, primarily purchased from data bureaus.
RFM: Classifying customers based on how recently they visited (Recency), how frequently they purchase (Frequency), and how much they spend when they do purchase (Monetary Value).
Customer Missions: A reflection of purchase intent, including what phase of the purchase cycle the customer is in (i.e. dreaming, researching, or searching with strong intent to buy).
Each of these segmentation approaches has unique strengths and an ability to answer a specific set of business questions. The trick is to consider segmentation not as a single monolithic model, but rather as a library of models that you can use alone or in any combination.
When you visit a library with the intent of learning about something, sometimes you need one book, and sometimes you need more than one book. That principle applies equally to the segmentation library. The grid below is a list of five common business challenges you might be trying to answer through segmentation, and the types of segmentation that will best address those questions.
When it comes to mass media, segmentation is helpful for identifying broad-scale customer characteristics often found in geo-demographic segmentation—income levels, household composition, and generation all come into play here. Mass media can also be influenced by attitudinal segmentation, where the perspectives of the target customer base are used to define a messaging strategy.
Digital media targeting tends to be more directly driven by specific behaviors exhibited by individual customers. This can include both directly tracked actions (e.g., abandoned a product in an online shopping cart) and inferred missions (e.g., frequent searches for product review could mean customers are in “research” mode).
Personalization can be supported through a number of types of segmentation schemes. A history of purchase behavior can be used to make better product recommendations. An understanding of their state in the customer lifecycle can be used to customize messaging and offers. RFM/LTV (lifetime value) models can ensure the best customers receive the highest levels of service. An identification of a customer’s mission can also ensure that the only messages relevant to that mission are shown.
Customer experience design is a bit of a larger challenge to tackle, and segmentation is also critical here. A combination of a behavioral understanding (what customers are actually doing), attitudinal profiling (how they feel about those various experiences), and how those perspectives differ among best customers (RFM/LTV) should all be used to ensure that customer experience efforts are focused on the right things. Experiences may also be customized based on where customers are in the customer lifecycle (e.g., customers who say they want to cancel a subscription service are often routed quickly to a customer service agent).
Finally, pricing and product strategy can be supported through various types of segmentation. Behaviors can tell you what pricing and product strategies have worked in the past. Geo-demographics can tell you who you are designing for, and RFM/CLV modeling can determine whether there are differences based on customer value. Finally, mission-based segmentation can tell you whether you have any product or pricing gaps based on what customers need states are.
Whatever your business question, remember that segmentation is not a single book, but rather an entire library at your disposal.
The Segmentation Library
BY BROOKE NIEMIEC – CHIEF MARKETING OFFICER
A while back, my colleague Chuck posted an article on how segmentation, in its many different forms, is everywhere. It’s everywhere because it’s useful—a form of classification that helps us categorize things in a meaningful way. It’s also something we all use in our personal lives.
Take a moment and think about your social network. My guess is that you probably know which friends you would invite to a movie, which ones you would take on a white-water rafting adventure, and which ones you would send a birthday gift to. I would also guess that those lists do not contain completely distinct groups of individuals—some of your friends might be members of several of those lists, and others may only fit into one.
Customer segmentation is no different. There are countless ways to create groups of customers based on similar characteristics. One “master” segmentation won’t cut it anymore—different types of segmentation are required to answer different business questions. But where do you start?
Below are six common segmentation types and a brief definition of each.
Behavioral: Tied to actual customer behavior, based on variables such as what customers buy, how frequently they visit, and how much they spend.
Attitudinal: Based on self-reported beliefs and attitudes, often about the brand, category, or lifestyle.
Lifecycle: Used for managing the entire customer journey with the brand over time, from acquisition to retention to reactivation.
Geo-demographics: Broadly applicable segments that capture customer data by geographic region and/or by socioeconomic factors such as gender, income level, or age group, primarily purchased from data bureaus.
RFM: Classifying customers based on how recently they visited (Recency), how frequently they purchase (Frequency), and how much they spend when they do purchase (Monetary Value).
Customer Missions: A reflection of purchase intent, including what phase of the purchase cycle the customer is in (i.e. dreaming, researching, or searching with strong intent to buy).
Each of these segmentation approaches has unique strengths and an ability to answer a specific set of business questions. The trick is to consider segmentation not as a single monolithic model, but rather as a library of models that you can use alone or in any combination.
When you visit a library with the intent of learning about something, sometimes you need one book, and sometimes you need more than one book. That principle applies equally to the segmentation library. The grid below is a list of five common business challenges you might be trying to answer through segmentation, and the types of segmentation that will best address those questions.
When it comes to mass media, segmentation is helpful for identifying broad-scale customer characteristics often found in geo-demographic segmentation—income levels, household composition, and generation all come into play here. Mass media can also be influenced by attitudinal segmentation, where the perspectives of the target customer base are used to define a messaging strategy.
Digital media targeting tends to be more directly driven by specific behaviors exhibited by individual customers. This can include both directly tracked actions (e.g., abandoned a product in an online shopping cart) and inferred missions (e.g., frequent searches for product review could mean customers are in “research” mode).
Personalization can be supported through a number of types of segmentation schemes. A history of purchase behavior can be used to make better product recommendations. An understanding of their state in the customer lifecycle can be used to customize messaging and offers. RFM/LTV (lifetime value) models can ensure the best customers receive the highest levels of service. An identification of a customer’s mission can also ensure that the only messages relevant to that mission are shown.
Customer experience design is a bit of a larger challenge to tackle, and segmentation is also critical here. A combination of a behavioral understanding (what customers are actually doing), attitudinal profiling (how they feel about those various experiences), and how those perspectives differ among best customers (RFM/LTV) should all be used to ensure that customer experience efforts are focused on the right things. Experiences may also be customized based on where customers are in the customer lifecycle (e.g., customers who say they want to cancel a subscription service are often routed quickly to a customer service agent).
Finally, pricing and product strategy can be supported through various types of segmentation. Behaviors can tell you what pricing and product strategies have worked in the past. Geo-demographics can tell you who you are designing for, and RFM/CLV modeling can determine whether there are differences based on customer value. Finally, mission-based segmentation can tell you whether you have any product or pricing gaps based on what customers need states are.
Whatever your business question, remember that segmentation is not a single book, but rather an entire library at your disposal.
Tags: