RFM Is Dead. Long Live RFEM


Share on LinkedIn

Ok, I admit it. I made the title of this post more dramatic than it needed to be just to engage you. And engagement is exactly what this thought is going to be about. Not engagement in the strategic sense that Graham Hill brilliantly discussed in an earlier post. Rather, just engagement in the sense that marketers have used RFM, namely for prioritizing customers for targeting with marketing communications.

Ok, Ok, so I am not talking about providing customers with a better experience. This one is purely for the marketer’s benefit. Namely to spend effort and money by directing marketing investments such as direct mail or e-mail to those most likely to respond. RFM metrics have been used for a long time for that purpose. RFM really isn’t dead at all. Recency is still a highly successful indicator as far as I know. Web analytics however can contribute more than just R, F, and M on each customer. Especially, with the advent of web 2.0 and more interactivity on web sites web analysts are trying to define E, i.e. engagement as a metric for scoring visitors.

So a visitor may be fairly R, moderately F, may be a small m rather than a M. But if the visitors has been a big E, then they should be more likely to be a good target.

This may fail in certain situations. If you want a silly example: The Victoria’s Secret web site has lots of visitors with a capital E, R, and F, but no M.

A.) I work for a software firm that provides web analytics solutions combined with modeling software that do this, namely at Unica. But one of my direct competitors, Webtrends, has just released a product by the name of Score for this very purpose too. And one of the best known names in web analytics, Eric Peterson, is all about engagement metrics. Our industry is trying to make sense of the practical and tactical use of this metric.

B.) I have not researched this in any way to confirm it. But if you think of an abandoned shopping cart follow up e-mail program, you essentially have an RfEm program right there. So the only thing that is new in this thought is to use a new acronym and retire the old one.



  1. Akin

    Another great post. Thank you.

    I agree with you. Marketers do need to know more about customers and their propensity to buy than Recency, Frequency & Monetary value alone can provide. Engagement is one possible way to do this (although it still lacks a robust definition and measurement framework). So is the Net Promoter Score (although it has been found to have serious methodological problems). So are p(Alive) models developed by Werner Reinartz at Insead.

    Clearly there is still plenty of work to do to understand customers, the jobs they are trying to do, the context in which they are trying to do them and what all this means for their propensity to buy.

    Your postings make a valuable contribution to this discussion.

    Graham Hill
    Independent CRM Consultant
    Interim CRM Manager

    PS. I have been careful to avoid reciprocation bias in this post. I am not sure it worked!


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here