Have you ever heard of the saying, “80 percent of your business comes from 20 percent of your customers?” Then you’ve heard of RFM analysis (even if it’s just informally!). Keep reading for an overview of what it is, when to use it, and everything else you might need to know.
RFM stands for Recency, Frequency, Monetary Value. Recency is the time since the last purchase. Frequency is the total number of purchases made. Monetary value is the total monetary value of the customer. RFM analysis is a customer segmentation technique.
It allows businesses to use past purchasing behavior to divide customers into groups in order to market to them most effectively. It utilizes the idea that not all your customers are equal. The people who spend the most money are the ones you want to market to more. The people who don’t spend a lot of money, or no money at all, aren’t ones you want to spend a lot of time marketing to. After all, they haven’t made purchases in the past, so they’re unlikely to suddenly change that behavior.
The benefits of RFM analysis include increased customer retention, response rate, conversion rate, and revenue. It’s been shown to not only make the company more money but to keep customers engaged and coming back for more.
Statistical approach – Traditional way
You can perform an RFM analysis by dividing your customers into x number of equal groups according to the distribution of values for recency, frequency, and monetary value. You can use the chart below, where the number of groups is selected to be four, to help you get started.
|Recency (R)||Frequency (F)||Monetary Value (M)|
|Quartile 1 (R=1)||Quartile 1 (F=1)||Quartile 1 (M=1)|
|Quartile 2 (R=2)||Quartile 2 (F=2)||Quartile 2 (M=2)|
|Quartile 3 (R=3)||Quartile 3 (F=3)||Quartile 3 (M=3)|
|Quartile 4 (R=4)||Quartile 4 (F=4)||Quartile 4 (M=4)|
If you’re looking to find your Best Customers, you’ll be looking at your most recently purchased (R=1), highest quantity (F=1), and most spent (M=1). Each of these combinations will tell you something different about your customer and also how you should act accordingly.
RFM 2.0 – Unsupervised clustering model
As seen above, the statistical model assumes the data to be evenly distributed, this is not always the case, or almost never. Instead, if having access to Machine learning technology, a Unsupervised clustering model could be used. In summary it uses the input from the three variables,
By executing millions of iterations it learns how to segment the data into a undefined number of segments where all segment are significantly different and includes customers that share the same characteristics. Keep reading below to see how you should utilize your results from your analysis.
Here’s another chart to help you see just how easily you can segment and market to customers and their buying behavior.
|Champions||Bought most recently, most often and spend the most||No price incentives. New products and loyal programs instead|
|Loyal Customers||Buy most frequently||R and M could be used to further segment and understand marketing strategies|
|New Customers||Customer bought recently, but are not frequent buyers||Run campaigns that focus on loyalty, like for example membership deals/benefits etc.|
|Potential Loyalist||Recently made a couple of purchase at above average frequency||Run campaigns that focus on loyalty, like for example membership deals/benefits etc.|
|Can’t Lose them||Made frequent and big purchase in the past, but it has been a while since the last purchase||Reactivation Campaigns|
|Needs Attention||Purchased recently but and have not made previous frequent purchases||Aggressive price incentives|
|At Risk||Haven’t purchased for a while, used to purchase frequently with decent spend||Aggressive price incentives and reactivation campaigns|
|About to Sleep||Haven’t purchased for a while, used to purchase frequently with decent spend||Aggressive price incentives and reactivation campaigns|
|Hibernating||Made small purchases frequently but a long time since last purchase||Don’t spend to much on marketing|
|Lost Spenders||Stopped buying, used to purchase below average on frequency, but spend good money||M could be used to further segment and find the once that should be targeted for re-acquiring|
|Lost Customers||Stopped buying, used to purchase below average on frequency, spend decent amount of money||M could be used to further segment and find the once that should be targeted for re-acquiring|
|Lost Cheap Customers||Last purchase long ago, purchased little, didn’t spend below average||Don’t spend a lot re-acquiring business|
|Dead Beats||Last purchase long ago, purchased little, spend the least||Don’t spend re-acquiring business|
This table shows marketing techniques that you can use once you’ve segmented your customers. You’ll need to use your customer data in order to do this. A simple RFM analysis tool can help you to sort through your customer purchase history data in order and evaluate it for this type of information.
Want to know more about RFM? Engage offers a tool for an RFM analysis, based on a unsupervised clustering model, that can help you up your game. Start appealing to those “Champions”, “Almost Lost”, and “Needs attention today”. You’ll be shocked at how effectively it works and helps to boost your business.