2010.06.24 16:57:27

John McDaniel

A question about an insurance company’s “free service” on the optcom list prompted a discussion about RFM market segmentation.  Since many questions were raised about RFM, I decided to draft a quick blog primer on the subject.  At the outset, I want to be clear that I believe this approach can have value in market segmentation for an optometric or other health care practice, but the logistics are a little daunting.  I will address a few of the issues I’ve encountered while attempting to implement RFM in optometric practice at the end of this blog entry.

What is RFM Market Segmentation?
The concept of RFM segmentation is not new and is rather straight forward.  There are three premises the concept is based on:

1.  Customers who have purchased recently (R) are more likely to purchase again, when compared to customers who have not purchased for a long time.

2.  Customers who purchase frequently (F) are more likely to purchase again when compared to customers who have purchased just once or twice before.

3.  Customers who have spent the most money (M) in total are more likely to purchase again.

These concepts are rather obvious.  But over time, the power of this notion of recency, frequency, and monetary level has been expanded into less obvious situations.  For example, even for someone who has never been your customer, you can predict the likelihood of them becoming a customer.  Assessing the recency of their visits to your website, the frequency of these visits, and money spent (even if it is zero) during those visits helps you predict the likelihood of them becoming your customer in the future.

In fact, the model has been used exclusively for non-customers where the monetary value is not utilized (since all non-customers have all spent zero money, the factor is of no value in rating the prospects).  This two factor, R and F, approach can be very powerful when determining which prospects should have marketing resources allocated toward attracting them.

The core idea of RFM (or RF) is to score customers/prospects on R, F, and M.  These scores are then compiled into a single value that is used to determine the relative worthiness of expending marketing resources on each individual customer/prospect.  There are many variations in scoring systems, and many specific corrections and adjustments that can be made to the process depending upon industry.  You will need to determine a cutoff score, then only allocate marketing resources to attract those with that score or higher.  This makes your marketing efforts both more efficient and effective.

What are the Pitfalls of RFM in Health Care Marketing?

Effective RFM needs one thing…data.  Lots and lots of data on who is accessing your website, when, and how often…who is calling your office, when, and how often…who is spending what, when, and how often…who is viewing your Facebook page, when, and how often…etc.  You get the idea.  The problem in optometric practices is that much of this data is either not collected at all, or it is collected in a form that is rather difficult to access and study.  If there is no data, there is no use for RFM.

Assuming you have enough accessible data, here is something else to consider: is the total market big enough to justify the analysis to further segment it?  An example may help explain this.  Let’s say a practice is one of two optometric offices in a smallish city of 50K people.  Assuming that the practice has about 10K active/semi-active patient files, that means 40K of the city's residents are not patients, or have not been patients for several years.  You also have to factor in the fact that the other practice will account for some of the 40K remaining people.  In case, let’s say that other practice will account for 10K folks.  Now, ask yourself, is it worth collecting the RFM data needed to segment that remaining 30K people to determine who you are going to prioritize your marketing efforts to attract?  Maybe.  Maybe not.  It depends upon what the marketing efforts are.  If it is a mass mailing effort that costs $0.50 a piece, you might be able to eliminate 10K from the mailing list.  That will save you $5K in costs.  But, how much will it cost you to access and analyze the RFM data and to score it?  Probably more than $5K.  Hopefully, you can see how this approach makes sense for more expensive per person efforts, but not for less expensive per person efforts.  In most optometric offices, the marketing activities are in the “less expensive” category.

As I said at the beginning of this blog, I do believe a form of RFM is valuable in the business of optometric market segmentation.  To be successful, the process requires a good data collection and retrieval system.  It also, likely, needs to focus on internal customers where that data collection is possible, at a very low cost.  Finally, the scoring system must be well crafted, weighted, and able to be incorporated into a solid practice management system (electronic PMS).  In other words, it has be easy to do and easy to access.  If it is not easy, it will never be used.

A final note – a large number of details have been omitted in this blog on RFM market segmentation.  There are text books dedicated to RFM and that level of detail is well beyond the scope of this blog.

Le Meas, 

John McDaniel, OD, MLHR
President/Founder
Waugoo Consulting Group, LLC
This e-mail address is being protected from spambots. You need JavaScript enabled to view it

RFM | marketing | market segementation