RFM Analysis: Why and How to Divide Customers?

RFM analysis is a method of working with the client base, which allows you to identify the most money customers. It isn’t difficult and helps find out who brings in the most money, and for whom the company just pours the advertising budget.

RFM Analysis Takes Into Account the Age, Quantity, and Amount of Purchases

Imagine that the online Beauty store sells its own natural cosmetics. They have 5,000 customers in their database, to whom they regularly send SMS and emails, and spend lots and lots of money. And the conversion, that is, the return on investment is small.

The online retailer starts to figure out to whom these messages go. It turns out that most of their customers have made one small purchase and never come back. But there’s a fraction of customers who regularly make large purchases. The store decides to text and write letters to the best customers, and reduce the cost for everyone else to a rare text message at Christmas.

As a result, Beauty reduced spending on one-time customers and concentrated on interesting offers for regular ones.

This example illustrates well how RFM analysis works. It helps segment all customers into different groups based on three characteristics:

  • Length of last purchase, R.
  • The number of purchases, F.
  • Amount of purchases, M.

RFM analysis can show that a company is spending a lot of money to attract a customer who bought just one item three years ago and hasn’t come back or played only demo casino table games online not trying real-money bets. And it doesn’t pay enough attention to a customer who recently made several purchases for different amounts and is interested in buying more.

Pros and Cons of RFM Analysis 

But no matter how good RFM analysis is, it’s important to understand that in the real world, not everything is perfect, and this tool has its pros and cons:

  • Savings. Helps reduce the cost of marketing campaigns by optimizing targeting.
  • Convenience. Suitable for online commerce, mailings, direct marketing, and nonprofits.
  • Combinability. It’s easy to combine with other customer relations tools.
  • Loyalty. Reduces negative customer behavior through controlled targeting.
  • Customer base. The effectiveness of the analysis depends on the company’s database (you can’t go far with 5 customers).
  • Cyclicality of purchases. Not suitable for companies with “one-time” clients or goods.
  • Transaction experience. RFM analysis is based on history. It shows the past and doesn’t predict the future.
  • The complexity of calculations. Without software and scripts, it is difficult to calculate RFM evaluations, especially for large companies.
  • Customer movement. The database is live, segments can change, and the analysis will have to be updated (at least once a year).

Gathering a Base for Analysis

Offload the Customer Base

RFM analysis can be used with a base of both 1,000 people and 15,000. The main thing is to have a base. The customer base for RFM analysis includes all customers who have made at least one purchase of any amount.

Define the Parameters of the Analysis

Next, analyze the base data in Excel and determine the main parameters. Each company will have its own indicators of age, frequency of purchases, and average check, but usually all entrepreneurs have this data.

And now an important point: we numbered each parameter with a number from 1 to 3. It stands for:

  • 1 is weak. Older customers, one-time shopping, low check.
  • 2 is medium. Average age, infrequent purchases, average check.
  • 3 is excellent. Recent and frequent purchases, high average check.

After that, you can automatically number each customer for each indicator and then collect a summary result.

Analyze the Result

So, we see that each client gets a different number for each parameter and the summary result as a whole.

This is RFM segmentation. From there you can analyze the results: for example, remove all the work with a result of 111, because it isn’t an effective client, and focus efforts on working primarily with threes and twos. Or on the contrary, think about why there are so many ones in the company, perhaps there is something wrong with the product or service.