RFM Analysis in E-commerce
Your strategy to grow makes sense. But are you also looking at your current customer base? With the RFM analysis, you divide your current customer base into different groups in order to serve them even better than you did before. This is called segmentation. There are several segmentation options, one of which is the RFM analysis based on customer value. Who are your best customers? Which customers have you lost? Which customers should you pay attention to? The RFM analysis provides answers to all these questions.
An RFM analysis is a simple and effective way to divide your customer base based on 3 values:
- R: Recency (how recently did a customer make a purchase)
- F: Frequency (how often does the customer buy)
- M: Monetary (Value, how much did the customer convert)
The RFM analysis is based on the marketing principle '80% of sales come from 20% of customers'. This is a fun fact, but which 20% of customers are these? In addition, the model assumes that people who have recently purchased something from you are much more likely to respond to an offer from you than they would be for someone where it has been a longer time since they ordered. Email is an important channel to use, as you capture all of this response data from your customers in your email system. With the insights of the RFM analysis, you can adjust your marketing activities. How do you start with an RFM analysis?
The beginning of your RFM analysis
You start at the beginning: you make an RFM parameter per point. With parameter, we mean that you provide the points with scores of 1, 2, 3, and 4 (with 4 being the highest). With this point system, your customers can be classified by point. These three scores together form the RFM value of your customer. Your very best customers are those with an RFM value of 4-4-4, your least valuable customers are those with a score of 1-1-1. Anyway, you have a convenient system ready for yourself. Fill it in! Don't panic, Excel has a huge number of ways to do this.
Step 1: create your data model
Create a data model in Excel of all transactions in a particular period and make sure you can determine the R, the F, and the M for each customer. In other words, how many days has it been since your customer last bought something (R)? How much did your customer buy at that time (F) and what was the turnover from this (M)? A handy CRM system makes a report of this in no time.
Step 2: Divide your customers into groups
Now that you have the data, it is easy to divide your customers into groups. You make this division for all values (R, F, and M) based on a score of 1 to 4. In this case, 1 is the least valuable and 4 the most valuable. Divide your scores logically: 1 could be a value that your customer ordered from you more than a year ago and 4 could be a value that your customer ordered from you just this week.
Step 3: Aggregate the scores for a total RFM score
As a final step, you're going to aggregate the R, F and M scores for each customer. This results in 64 segments, with the 444 segments representing the most profitable customer to 111 representing customers who are the least valuable. In the image below, you can see an example of how we created this, where the green customer is a valuable customer and the red customer is a less valuable customer:
Making your RFM analysis transparent
Now that you know how much a customer converts, how often they order and how much time there is between certain orders, you can act on this. The next step is to divide your RFM scores into customer groups.
With the example above, you see that customer 245344 belongs to your top customers and customer 3345454 to your flop customers.
Now create logical customer groups based on the total RFM score, such as:
444 - Top Customers. These customers score high on all three points. These are your customers who bring in the most sales.
111 - Flop customers. These customers score low on all three points. These are your customers who provide the least amount of sales
441 - Many small purchases, This customer group regularly orders from you, but the turnover remains low. This target group you ultimately want to push to a higher-order amount per order.
144 - Inactive tops. These customers purchased frequently in the past and their customer lifetime value is high. You can see that this customer group is starting to become inactive.
414 -Promising customers. They bought a few days ago and spent a lot. This group usually contains customers with a potentially high customer lifetime value. Develop your customer relationship with these so that this group will order from you more often.
Once you have created logical customer groups, it is time to make this visual for yourself. This can be useful, so you can physically put your customers in a “group” and physically move them to another group. For example:
In the example above, recency is on the x-axis, and frequency-monetary value is on the y-axis. The percentages indicate how many customers relative to the total number of customers are in that segment. This is also how you make it transparent to yourself. You want to move your customers from the bottom left to the top right with a targeted marketing strategy.
Using your RFM analysis
Now that you've mapped out what stage each customer is in, it's possible to act on this. Now the idea is to use each customer group as the basis for your growth strategy. Below are some useful strategies for each segment:
Lots of small purchases
In this segment, you want to get them to make larger purchases so that their lifetime purchase value increases. Larger purchases can be generated by offering more cross- and upsell items.
Inactive buyers
These customers are the ones you want to get to order something more often than they feel necessary. Create the urgency that this group should order something now. You can do this with temporary offers or temporary discount codes.
Flop customers
Your flop customers do not score well on anything: not on sales, not on the frequency, and not on recency. This is a difficult group to reach. Start your approach small: indicate to this group exactly why they should order from you. Maybe they are not aware of your benefits (such as customer service, delivery times, shipping costs, etc.)
Loyal customers
Your loyal customers are your customers who score well on everything, but could score even better on recency. Reward your loyal customers occasionally that they order from you more frequently. A personal e-mail, a savings program or a free product: rewarding your customer is definitely worth it.
Needs attention
Customers who need specific attention in order to proceed to sale are approached differently again. In this case, marketing automation is indispensable. Set up an email flow with touchpoints when you want to get back in touch with your customer. Do you see that the repeat purchase is at 4 weeks? Then you can come to your customer's attention after 3 weeks to order a product again.
Customers I'm going to lose
These customers need an extra push to make a repeat purchase from you. Do they not make this repeat purchase? Then they've lost track of you. Give these customers a discount code. Do everything you can to not lose your customer. A personal touch in communication certainly can't hurt.
Promising customers
The only thing missing from this group is frequent ordering. Use marketing automation to cleverly target this group and build a flow of e-mails around it. The power of repetition is very important here.
Potentially loyal customers
This group spends a lot in your webshop and recently ordered something from you. This group can also order from you more often, so they become even more loyal. Maybe this group lacks a personalized approach. This group is very suitable for cross-selling and upsell campaigns so that they order more products and more frequently.
New customers
Finally, you have new customers. These are customers who are not at all familiar with your shop, your offer and your advantages, but have perhaps already placed an order with you. Build an email flow around these new customers: introduce yourself as a shop, what do you offer and at what price? What benefits can they expect from you? The more your customer becomes familiar with you, the more likely it is that this group will consistently order from you.
The translation of RFM analysis to e-mail
In this article, we covered how to create an RFM analysis based on your customers. But did you know that you can also create an RFM analysis based on your email numbers? For example, you can determine an engagement RFM score, in other words, an eFRM, for each newsletter subscriber in your email database. You then determine the three values slightly differently:
- R: Recency (based on the most recent date the recipient opened your email or clicked on something)
- F: Frequency (total number of times the recipient has opened or clicked on an email)
- M: Monetary (how much did the customer convert)
In this way, you determine the engagement of your customers and see, based on your email marketing, to what extent your target audience is interested in certain emails.
So you see again that the RFM analysis can be used for different purposes. The RFM analysis is one of the methods a marketer has at his disposal to identify customers. There are plenty of other analyses you can perform on your data. Good luck!
