Statistics Definitions > RFM (Customer Value)
RFM and Customer ValueRFM stands for “Recency, Frequency, Monetary” and is a way to figure out who your most valuable customers are. For example, a customer who spent $1,000 three times in the last month is a lot more valuable than a customer who spent $100 once in February of last year. RFM is based on the Pareto Principle (the 80/20 rule) that tells you 80% of your income comes from 20% of your customers. You use RFM to identify that top 20% of customers and focus your marketing on that market segment.
- Recency: when was the last time the customer made a purchase? Customers who purchased from you recently are more likely to buy from you again than customers from the distant past. This is the important ranking factor–that’s why it’s first in the list.
- Frequency: how often does the customer spend money? A customer who is in every day is much more likely to buy again that someone who only comes in once a year.
- Monetary: how much did the customer spend? A customer who makes a large purchase is more likely to buy again than a customer who spends a lot less.
With this knowledge, you could probably assign your own points system, giving more points:
- To the people who made a purchase recently than those who purchased from you years ago.
- To the highest spenders than whose people who spend a lot less.
- To your repeat customers rather than the one-time customer.
In fact, this informal way of calculating RFM has worked for many direct mailing marketers for decades. You could even use the original customer data to rank customers (i.e. from largest dollar amount spent to least).
A more formal way of assigning points is to assign a number from 1 to 5 for each category, where 5 is the highest. In order to calculate RFM you’ll need some data on your customers:
- Their most recent purchase date.
- Number of purchases within a set time period (i.e. one year).
- Total sales from that customer (you could also use average sales or average margin).
Next, decide on categories and assign numbers for those categories. Your exact categories will depend on your type of business, how long your business has been open and other factors, but as an example:
- Recency: within a week (5), a month (4), a quarter (3) six months (2) a year (1)
- Frequency: 10 purchases in the last year (5), 8 purchases (4), 6 purchases (3), 4 purchases (2), 3 purchases or under (1)
- Total sales: Over $10,000 (5), $8,000-$9,999 (4), $6,000-$7,999 (3), $4,000-$5,999 (2) $3,999 or less (1).
Using the above scale, a person who spend $10,999 twice in the last six months with their last purchase on Monday of this week would have an RFM score of 515.
With this scoring system (i.e. 1 through 5 for each category), you’ll end up with 125 possible scores. In general, the higher the score, the more valuable the customer.
LRFM adds a fourth variable to the mix: length of relationship with the customer. For example, someone who has been a customer of yours for ten years is more likely to remain a customer than someone who has only been around the last few weeks. LFRM has been used in a wide ranging variety of situations including healthcare patient analysis (for example, see Wu et. al).
Be careful not to over-saturate your best customers with marketing materials. By the same token, don’t completely ignore your lowest ranking customers. The customers with the lowest ranking RFMs should be seen as an opportunity for improvement.
Wei JT, Lin SY, Wu HH. A review of the application of RFM model. African Journal of Business Management. 2010;4(19):4199–4206
Need help with a homework or test question? With Chegg Study, you can get step-by-step solutions to your questions from an expert in the field. If you'd rather get 1:1 study help, Chegg Tutors offers 30 minutes of free tutoring to new users, so you can try them out before committing to a subscription.
If you prefer an online interactive environment to learn R and statistics, this free R Tutorial by Datacamp is a great way to get started. If you're are somewhat comfortable with R and are interested in going deeper into Statistics, try this Statistics with R track.
Comments? Need to post a correction? Please post a comment on our Facebook page.