{{ searchResult.published_at | date:'d MMMM yyyy' }}

Loading ...
Loading ...

Enter a search term such as “mobile analytics” or browse our content using the filters above.

No_results

That’s not only a poor Scrabble score but we also couldn’t find any results matching “”.
Check your spelling or try broadening your search.

Logo_distressed

Sorry about this, there is a problem with our search at the moment.
Please try again later.

RFM (Recency, Frequency, Monetary value) segmentation is one of the most tried and tested methods of segmentation used in direct marketing. It is based on the presumption that someone’s future actions are best predicted by their past ones.

So how can this popular method of segmentation be used in today’s data rich world, can it help answer the 64 million dollar question in email marketing: who do I send what to and when do I send it?

In the early days of offline direct marketing, RFM was often used to target consumers with expensive direct mail. It saved money and by targeting those most likely to buy again whilst removing those least likely to purchase; you would achieve a far higher ROI than if you mailed everyone. Its other benefit is that it is easy to apply and easy to understand. 

It lost favour in the early days of email marketing; because email was cheap to send, so therefore the segmentation was easy, “there’s one list, let’s send it to everyone, isn’t email marketing great!”

This approach worked for a while, but as time has passed this strategy has gradually delivered diminishing returns for email. This is due to the wave of change that developed within the modern online society, making the use of the media extremely difficult without proper targeting.

The consumer now expects to see only what they want and is becoming increasingly intolerant when that doesn’t happen.

This means that modern segmentation is not just about saving money, we segment now for relevance, and to increase the consumer’s engagement with the brand. This means that a marketing communication should be seen as a key customer touch point, and should be a positive experience not only for those that buy now, but also those that don’t (but might in the future).

I’m not saying that each email is a “moment of truth” situation, but if we cause subscribers to complain or unsubscribe we lose the ability to influence them in the future.

So we need to segment, and it’s more important now in the online world, than it has ever been before.  

In the original development of RFM segmentation, marketers used the data they had, not only the RFM but also category data as well. With this you could get to the who and what in a reasonably straight forward way.

RFM was one of the first “engagement” segmentation models, if you were in the top segments for Recency, Frequency and Monetary Value you were a highly engaged customer, and the other side of the segments, meant you were the most disengaged of the whole customer list!

So how does RFM cut it in the online world, and do the same rules need to apply?

I think the answer is yes, but let’s do a bit of redefining first. We need to accept that engagement data now goes beyond the transactional. So, by dipping into the rich sauce of online behavioural data, maybe we need to call it what it is; Engagement RFM (e-RFM).

This means that we look at what people have done in the past and what they are doing now in real time. Are they interacting with email (Recency, Frequency)? When did they last visit the website (Recency, Frequency, Duration)? What are they like as a customer (RFM)?  The element we are adding to RFM (showing us likelihood to purchase) is the web behavioural data (which can show us “intent” to purchase).   

Online RFM allows you to be more accurate with your predictions, by combining all the key behavioural data. This gives the marketer the control they need to make informed decisions on segmentation and targeting.

Developing this type of segmentation is a bit like eating an Elephant (apologies to the WWF), it’s only possible in small chunks. You should start by looking at someone’s RFM score or how recently they have interacted with you online when choosing your most engaged segments for example. But the important thing to remember is that it doesn’t stop there and by combining more of the key data points, your targeting can become increasingly accurate and meaningful.

Engagement RFM provides a mechanism to successfully target people based on their past behaviour and their current position in the Customer Lifecycle.

One example of this would be the use of RFM in the management of sending frequency. Those who are the most engaged, would be sent emails at the greatest frequency (to influence conversion), whereas those who are less engaged, would be sent emails at a lesser frequency (to maintain engagement with the brand).  

The objectives for the different types of communication would not only affect frequency, but define the content and the message of each of the emails too.  

Ultimately, the goal of sending emails to people who want them, making them happy (rather than upsetting them) and influencing them to buy more, becomes a little closer to reality

Tim Roe

Published 10 January, 2011 by Tim Roe

Tim Roe is Director of Data and Deliverability at Redeye International and a contributor to Econsultancy. Follow him on Twitter, Google+ or connect via LinkedIn

22 more posts from this author

Comments (8)

Comment
No-profile-pic
Save or Cancel
Avatar-blank-50x50

John-Scott Dixon

Love seeing RFM discussions and think you are right on. I recommend checking out MailChimp for an automated engagement RFM - they rank each person in you list from 1 to 5 stars based on engagement. Then you can select which level or levels you'd like to address with a campaign.

almost 6 years ago

Tim Roe

Tim Roe, Deliverability and Compliance Director at RedEyeEnterprise

Thanks John-Scott That’s certainly a good way of flagging engagement; we’ve used a similar flagging system for a couple of years now, that takes into account time on database as well as response Recency (or not, if they haven’t done anything). I think the key element is to make it as usable as possible and this is what e-RFM tries to achieve and using it to drive automated behavioural emails, is certainly powerful stuff!

almost 6 years ago

Avatar-blank-50x50

Andrew Bonar

Great article Tim... I think it is important to try and remember the lessons that were learned in traditional Direct Marketing and whilst embracing the fact that offline and online/digital Direct Marketing have their differences, essentially they are still the same beast DM. We may need to rework some definitions and expand on the old models, but essentially what was true then in DM will be true today, if only we can identify how to leverage those lessons. I think you have put an excellent case together showing how RFM translates to e-RFM and agree wholeheartedly it is key to answering the $64 million dollar question. If the data can be harnessed and the right flagging systems are in place without a doubt it is extremely powerful stuff.

almost 6 years ago

Ashley Friedlein

Ashley Friedlein, Founder, Econsultancy & President, Centaur Marketing at Econsultancy, Centaur MarketingStaff

Great post Tim. I'm quite a fan of RFM but you're right that online brings additional data and dimensions, specifically engagement, which it makes sense to use. 

almost 6 years ago

Avatar-blank-50x50

Email Vendor Selection

Tim, You are really on to something here, but your saying RFM is one of the first engagement metrics isn't correct. RFM values in the old days where based on purchase behavior. You can purchase a lot and not be engaged, as well as the other way around and actually monetary value has nothing to do with it. It is added to the calculation to be able to find your top tier customers to pursue commercially. Still, behavioral data is one of the best predictors of future behavior, a valuable resource often not used by (email) marketers. Your new e-RFM is something more marketeers should have a look at. greetings Jordie van Rijn

almost 6 years ago

Tim Roe

Tim Roe, Deliverability and Compliance Director at RedEyeEnterprise

Hi Jordie. Thanks for your feedback and some good points, although I would consider a recent, frequent, customer, an engaged one (especially if RFM was all you had to go on). But as engagement has as many definitions, as there are people with opinions, we may need to agree to differ at the moment. I like your point about the different elements of RFM, have differing uses (as your observation on the monetary value pointed out) and for differing business models, segmentation such as FRAC(L) maybe more useful (Frequency, Recency, Amount, Category, Loyalty). The same goes for what defines engagement online and it is important to do the data analysis groundwork, so see what hits the spot for your particular business.

almost 6 years ago

Avatar-blank-50x50

Brian Curran

Great Post Tim.....it's the principle that is the key here. Online brings lots of additional data which can only improve the consumer relevance of your approach to RFM. In the old days, you used whatever you had in terms of engagement, purchase and anything else you could get your hands on. This is as true now except we have lots more to play with as well as much more event driven & behavioural elements to overlay(abandoned basket and browsing emails for example). More folk with a bit of practical experience of building segmentations and thinking about the customer in this way is a good thing which too few are actually spending time on...don't leave it all to your ESP as you'll end up asking them who your customer is which is something you might need to know yourself I would suppose...

almost 6 years ago

Matt Clarke

Matt Clarke, Ecommerce Director at B2B

Very interesting. I'm actually just in the process of working on a project to introduce RFM to our email marketing - we're hoping it's going to give us big rewards.

I'm working towards making ours segment-specific, so we can define RFM requirements for individual product categories and score customers against those categories, as well as overall. This should give us a range of RFM scores per customer, which will hopefully identify which product categories they'd be most likely to purchase from, which means we can target the right items at them.

almost 6 years ago

Comment
No-profile-pic
Save or Cancel
Daily_pulse_signup_wide

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Daily Pulse newsletter. Each weekday, you ll receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.