Questions? (877) 643-7244
FutureNow Post
Friday, Jan. 30, 2009 at 6:05 am

The Power of RFM

By Bryan Eisenberg
January 30th, 2009

By now, many online retailers should be familiar with the abbreviation “RFM,” which stands for recency, frequency, and monetary value. For a refresher, here’s my explanation from 2002.

Over the past several columns, I’ve examined conversion rate basics. This week, we continue our study of the basics with an updated look at RFM.

Recency represents the number of days since the customer last completed the action you’re profiling. Frequency represents the number of times the customer has completed this action since the first time she completed it. Monetary value represents the total value (usually total sales) the customer created by completing these actions.

The classic RFM model produces scores that rank customers relative to each other for the likelihood that they will repeat the action being profiled. Any action can be profiled: visits, purchases, logins, and so on. High likelihood to repeat an action, providing this action has economic value to the company, means high future value. Low likelihood to repeat means low future value. RFM is that simple.

RFM is a commonsense way of sorting marketing and optimization decisions based on what your visitors actually do and what they spend.

What Can RFM Do for You?

Let me defer that answer to a friend, optimization junkie, and fan of RFM. According to Frank Malsbenden, VP/GM of Vision Retailing, parent company of Shoeline.com:

    The use of RFM metrics can drastically improve the effectiveness of e-mail marketing. It’s no secret that your best chance at repeat business is within 30 days of the customer’s last purchase. It’s not hard to grasp that frequency of purchases are a key metric in understanding the “health” of your customer database. And if you think about what you expect for service levels at retail establishments you spend heavily with, it’s easy to put yourself in the shoes of customers who spend big bucks with you. This is RFM, and catalog empires were built on the premise that recency, frequency, and monetary metrics should be the driving force in customer retention strategies…

    But sometimes in e-commerce, we outthink ourselves, get too cute by half, or just get too plain confused by all the data that flies our way on a daily basis. Instead of keeping it simple, we get too “sophisticated.” What we should be doing is mastering techniques like RFM that have been proven over time. The possibilities with RFM and e-mail are endless. At Shoeline.com, we’ve started off with the basics. We developed e-mail templates for customer’s who purchased 30, 60, 90, 180, and 365 days ago. The content for each template is dynamically driven based on the RFM score of the recipient as well as references to previous purchases. We’ve developed the process so that these e-mails are sent automatically every day. All we have to do is change the content for the templates on a seasonal basis to ensure thematic relevance. The results speak for themselves. In 2008 the difference in open-conversion rate (orders/opens) was 31 percent higher for RFM-driven e-mails versus our traditionally segmented e-mails. The difference in straight conversion rate (orders/recipients) is so stark I’m afraid to go public, for fear of losing credibility. These results have prompted us to increase the percentage of RFM-based e-mails in 2009, and you can be assured that someday soon, 100 percent of our e-mails will be RFM-based.

    Obviously the value of RFM goes beyond e-mail and can be used to increase SEM efficiency, ad targeting, and even merchandising. Still the simplest and most profitable use of RFM scoring is to identify and resell your best customers over and over.

Getting Started With RFM

Start by reading my previous columns on the subject: “Betting the Farm on RFM, Part 1” and Betting the Farm on RFM, Part 2.

Once you grasp RFM fundamentals, you’ll be inspired. Once you sort by these criteria, you’ll quickly find new and exciting ways to use them, such as:

  • Is that PPC (define) campaign really bringing in high RFM customers or just low RFM customers? Any campaigns you would kill?
  • What do high-scoring customers buy more of? Are there any patterns? Can you make them an offer to sell them again?
  • What other behaviors are high RFM visitors engaging in on your site? Are average number of pages higher or lower? Is time on site higher? And so on.
  • Create new KPIs (define) designed to optimize higher RFM visitors retention.
  • Does RFM data give you any insight on merchandising and inventory?

My one tip when using RFM: don’t waste too many resources turning low RFMs into higher ones. It’s much more efficient to keep higher RFMs engaged.

If you need a quick immersion in RFM, again I highly suggest “Drilling Down: Turning Customer Data into Profits with a Spreadsheet” by good friend Jim Novo.

Have you learned anything interesting from employing RFM techniques? Let’s us know in the comments section what you’ve found.

Add Your Comments

Comments (18)

  1. Hi Bryan, good post!
    I love the RFM power, altough I couldnt yet find a good way to use it on SEM campaigns, nor to get important insights from Web Analytics tools.
    The main reason is due to the lack of personal information from customer / visitors. So, it’s quiet difficult to use this RFM data to use it then on a email platform. ¿Any suggestions?

    Rgds.

    JC

  2. Anyone have thoughts on RFM with subscription B2B sales? Would it simply be up-sells and cross-sells?

  3. I can testify to the power of doing an RFM analysis on PPC campaigns in particular. I analyzed specific Adwords groups for one company, and realized they had many keywords that provided good ROI on the front end, but these customers would rarely buy again.

    On the flip side, we discovered keyword groups that had mediocre front end ROI, but excellent lifetime value. Based on these findings, we were able to decrease or increase funding toward these keyword groups.

  4. Two additional dimensions that prove very helpful are V (variety) and T (tenure). Variety doesn’t have to be limited to purchased products–you can apply it to content consumed as well.

  5. RFM is very powerful, but you really should experiment. Do a test campaign and then look at you response rates for each dimension of RFM. Some businesses will find their marketing efforts will be best suited by using FRM or MRF as examples. Plotting your response rates by quintiles for each dimension will tell you if your customer base requires a non-standard RFM approach. More can be found in a whitepaper at, http://www.staffordsbsg.com/RFM-Case-Study.html

  6. [...] Segment your email list. Don’t treat everyone like they are the [...]

  7. “I analyzed specific Adwords groups for one company, and realized they had many keywords that provided good ROI on the front end, but these customers would rarely buy again”

    Correct. I have come to the same conclusion when testing various PPC campaigns.

    BTW…Great post Bryan!

  8. I’m actually more familiar with RTFM. ;) More on the customer service side of things.

  9. I think that PPC is only for experienced marketers.

    If you do not know the game, you can loose a lot of money very fast.

  10. Critic for RFM:

    The method is descriptive only, and does not provide a mechanism to forecast behavior as a predictive model might. and when used to target customers for promotion, it assumes that customers are likely to continue behaving in the same manner.

    That is, it does not take into account the impact of life stage or life cycle transitions on likelihood of response.

  11. The only problem is that this not always works out that well when you apply it to other countries

  12. I agree with perros, this model won’t work that well for other countries.

  13. I’m familiar with SEM but RFM sounds new to my ear but you did a great sharing and posting this information. Thanks a lot buddy.

  14. I can testify to the power of doing an RFM analysis on PPC campaigns in particular. I analyzed specific Adwords groups for one company, and realized they had many keywords that provided good ROI on the front end, but these customers would rarely buy again.

  15. The only problem is that this not always works out that well when you apply it to other countries

  16. Thus RFM is better than SEM, SMM or any SEO technique?

  17. ihad to laugh at myself when I found myself in this exact situation. I had my boxing gloves on and I was ready to duke it out, expecting the worst.

  18. On the flip side, we discovered keyword groups that had mediocre front end ROI, but excellent lifetime value. Based on these findings, we were able to decrease or increase funding toward these keyword groups.

Add Your Comments

 

Print this Article
Share

Bryan Eisenberg, founder of FutureNow, is a professional marketing speaker and the co-author of New York Times and Wall Street Journal bestselling books Call to Action and Waiting For Your Cat to Bark and Always Be Testing. You can friend him on Facebook or Twitter.

More articles from Bryan Eisenberg

Marketing Optimization Blog
FREE Newsletter Sign-Up
send it once every: