How to Conduct and Perfect an RFM Analysis with Survey Research

How to Conduct and Perfect an RFM Analysis with Survey Research

An RFM analysis continues to remain relevant in the present day of big data and automation, despite being born in an age where direct mail was the most effective method of communication. 

Although its origin traces back to 1995, the RFM analysis is still a valuable method for performing customer segmentation, building customer personas, improving marketing efficiency and executing other applications.

Thus, market researchers should develop an analysis on the RFM model, as part of their market research techniques and to understand the value of various customers.

This article explores the RFM analysis, its importance, scoring system and model, how to conduct it and how survey research completes the process. 

Understanding the RFM Analysis

An acronym for recency, frequency and monetary value, RFM is a model for customer behavior segmentation. Also called RFM segmentation, this technique is used to analyze and estimate the value of a customer based on the three data points in its abbreviated title. 

The idea underpinning this kind of analysis is to segment customers based on the three major factors that make up customer buying behavior. 

This way, market researchers and business owners can identify which customers are regulars, big spenders and those who make one-time purchases. This kind of analysis allows businesses to distinguish between their different customers in this way, but on a larger scale

The RFM analysis model assigns each customer numerical scores based on the three measures to provide an objective analysis of their value to a company.

The system uses a scoring system that allows businesses to identify 64 kinds of customers (more on that in a below section). The RFM system allows market researchers to segment customers by assigning them scores based on the three data points of recency, frequency and monetary value. 

These scores, in turn, allow businesses to understand the value of their customers, whether they are worth pursuing and nurturing and how to better engage them. 

This analysis is based on the marketing adage stating that 80% of business comes from 20% of customers, which is also known as the 80-20 rule. 

As such, the RFM segmentation model provides a quantitative method for gaining a meaningful impression of customers. 

Recency, Frequency and Monetary Value

Marketers and market researchers should be well-acquainted with the three data points comprising the core of the RFM model. All three of these measures are proven to be effective predictors of the various personas’ willingness to engage with the messages in different marketing campaigns.

The following explains the meaning behind each measure and their unique KPIs:


This measure deals with the following: When was the customer’s last transaction? When was the customer’s last engagement?

Recency is usually expressed in days from the last purchase as its primary metric. However, depending on the product, it may be measured in hours, weeks or years.

Customers who made a purchase recently are more likely to have a product and its brand on their minds and are also more likely to purchase or use the product again. 

The KPIs of recency include:

  1. Date of the customer’s last purchase
  2. Date of the customer’s last engagement (ex: site visits, conversation with reps)
  3. Date of the customer’s last activity (ex: logins, in-app usage, commenting)


Frequency is concerned with the following: How often did a customer make a purchase within a given period of time? How often did a customer visit a store?

Aside from the physical space, this measure is also used to rate activity in the digital space, such as page visits, unique number of sessions, total logged-in time, time spent on site, time spent on a page, etc.

Customers who purchased once are more inclined to purchase again. First-time customers are good targets for follow-up advertising to retain them and convert them into frequent customers.

The KPIs of frequency include:

  1. The number of sessions/ visits 
  2. The number of click-throughs 
  3. The number of conversions

Monetary Value

This value answers the following: How much money did the customer spend within a given time period? What is the total revenue generated from the customer?

Not all businesses are ecommerce, such as news sites or content sites (blogs, social media, etc.); therefore, they cannot measure monetary value much like a traditional business would. Instead, they can assign an engagement metric that they deem valuable.  

Customers who spend a lot of money yield the most profit for a business. They are also more inclined to spend money in the future, more so than those who have spent less. These kinds of customers bring a high value to a business.

The KPIs of monetary value include:

  1. Total Revenue 
  2. Average Order Value (AOV) 
  3. Engagement Metrics- Useful for two-sided business models that don't directly sell products.

The importance of Conducting an RFM Analysis

An RFM analysis is crucial for businesses to conduct for a number of reasons. Firstly, it informs businesses on critical customer behaviors and allows them to build customer personas by ranking and grouping customers quantitatively. 

As aforementioned, this kind of analysis helps businesses predict how willing customers will be to engage in marketing messages and new offers. However, this kind of analysis can help with several other fronts. 

Conducting an RFM analysis gives businesses the opportunity to increase their sales. This is because using this data for decision-making allows them to understand how customers feel, think and shop, and most importantly, what fuels their buying decisions.

The RFM model allows businesses to support personalization efforts, which includes creating more personalized marketing campaigns or website and in-app experiences for logged-in customers. This in turn increases engagement, as it provides relevant messaging and offers to the current customer persona or group. 

This helps improve upon conducting a cohort analysis along with increasing retention. In terms of the former, an RFM analysis helps provide more info of customer behavior within different cohorts. As for the latter, personalization allows companies to provide more tailored experiences and messaging — the kind that can resonate more strongly with customers, thereby retaining them, either via engagement or keeping the business on their minds. 

Finally, RFM segmentation plays a role in customer retention, which is of the essence for any business. Although customer acquisition is important, retention carries even more weight for businesses, as existing customers are 50% more likely to try a business’s new product. In addition, 9% of businesses lose customers when they don't take customer retention seriously.

When products, services and experiences are personalized and engaging, customers are more likely to remain patronizing a business. Thus, the RFM model contributes to customer loyalty.

The RFM Model and Scores

The RFM model segments customers based on a scoring system. 

Each data point — recency, frequency and monetary value — is typically assigned a score of 1 to 5. 1 is the lowest score, signaling a poor ranking of the data point, while 5 is the highest and signals the most positive ranking.

An RFM cell is the collection of the three values for each customer. Companies can average these values together, then sort their customers from highest to lowest to identify the value of each. 

Using this scale, each customer can have a score from 111 to 555, with a total of 64 possible combinations, or 64 customer personas from just three points of data.

The scaling systems used in different companies that perform an RFM analysis will differ. Some will use a scale of 1-4, while others may elongate it to 1-10.

Examples of Personas Based on Their RFM Scores

Given the numerous customer personas that businesses can uncover after performing an RDM analysis, there are several key personas they ought to understand. These personas have critical consequences for businesses.

The following lists several major customer personas that an RFM segmentation brings to light, along with the most appropriate marketing campaign to target them with:

The Brand Champion (R=5, F=5, M=5)

This persona represents the ideal customer, as it exhibits the highest possible RFM cell, which is a representation of the 3 scores. When customers are correctly nurtured, they can become brand champions, spreading positive feedback about a business, thereby bringing more customers to a business themselves.

Marketing campaigns for this persona: exclusive offers, pre-purchase of new products, premium customer support, refer-a-friend bonus


The Loyal Customer (R=4, F=4, M=3)

Also a top-tier customer, as this persona visits or engages with a business frequently and has recently bought from them. Although their monetary value is in the mid-range, they can become a loyal customer — with the proper offers. There are several things marketers can do to increase their customer lifetime value or CLV.

Marketing campaigns for this persona: Loyalty campaigns, volume discounts, brand messaging showing the effectiveness of a product


The Possibly Alienated Customer (R=1/2, F=3/4)

This customer has been a regular at some point but has stopped buying recently, revealing a possible alienation. Their needs may have changed or they may have had a poor customer experience. Brands should nurture them back into becoming regulars, as a low recency score can quickly affect their frequency permanently.

Marketing campaigns for this persona: Customer satisfaction surveys, welcome back offers, more ads on their typical products/ services


The New Customer (R=4, F=1)

This persona has recently discovered a business and has engaged or purchased from it. They may also have been customers at some point, but stopped doing business for an interim. Businesses should foster relationship-building with this kind of customer, as they are new or possibly made a one-off purchase after an interim of no purchases/engagement.

Marketing campaigns for this persona: Email list sign-ups, introductory offers, hints, tips, useful content, social media offers


The One-Off Big Spender (R=1, F=1, M=4)

This kind of persona spends a lot of money on a one-time purchase, but then doesn’t return to the business, or does very rarely. This points to a specific need at a specific time, as they made a significant purchase, but only once. Marketers can target this persona by sending them messaging that is hyper-focused on their needs and interests. 

Marketing campaigns for this persona: Upgrade and maintenance offers, new promotions, content based on their interests, surveys


The Expired Lead (R=1, F=1, M=1)

The lowest-scoring customers and least viable kinds of customers, they don’t have a significant purchase history with a business and also score low on recent interactions. Businesses should not focus marketing campaigns on this kind of customer, as they are unlikely to bring any value to the company. These are also the weakest to turn into semi-regular customers.

Marketing campaigns for this persona: Awareness-stage messaging, content marketing, automated emails, along with some of the ads used for more valuable customers.

How to Conduct an RFM Analysis with Survey Research

When it comes to performing such an analysis, survey research is not a secondary task. It is a crucial part of identifying each customer via the RFM model. This is because surveys allow market researchers to ask their target market virtually any question.

The key is having access to an online survey platform that offers deployment across all geographical areas in which your customers reside, and ensuring that only the targeted population takes part in the survey.

Given that the RFM model only uses three data points, the bulk of the survey is going to be the three questions on those measures. When conducting such a survey, businesses can also execute competitive research — by asking these questions in relation to their competitors.

This can also help obtain a general idea of how customers shop. Researchers can do so by asking their customers about recency, frequency and monetary habits in the general sense, that is, without mentioning their brand. 

As such, they can also ask supporting questions to better understand each of the RFM data points.

The following lists major and supporting question examples for an RFM analysis in the general sense: 

  1. RECENCY: When was the last time you bought shoes?
    1. Supporting questions: 
    2. When do you intend on buying shoes again?
    3. Are you looking to buy shoes anytime soon?
  1. FREQUENCY: How often do you buy shoes?
    1. Supporting questions: 
    2. How often do you buy shoes in the winter?
    3. Do you intend to buy more shoes in the coming month?
    4. Is there a specific retailer or brand you prefer, or do you buy from several?
  1. MONETARY VALUE: How much do you typically spend on shoes?
    1. Supporting questions: 
    2. Are you willing to pay other amounts for different kinds of shoes?
    3. Which brands do you typically buy from?
    4. Is price an important factor when picking out different shoes?

Surveys provide a practical means of asking all of these questions and more. With advanced skip logic, brands can route respondents to follow-up questions based on the answer a respondent gave to a preliminary question.

Finding the Most Valuable Customers

An RFM analysis helps you to define some critical customer personas based on three major customer behaviors. But without an online survey platform, performing such an analysis becomes an almost impossible feat.

The correct online survey platform will enable businesses to reach a wide network of customers in their target market, ask them any type of questions and use artificial intelligence to stamp out low-quality answers and survey fraud

It will also help market researchers hone in on their RFM-based customer personas with additional data. Thus, businesses ought to choose the proper online survey platform to conduct an effective RFM analysis.