Category Archives: Analytical Culture

New RFM: Using RF or RM Instead of RFM

Analytical Culture, Customer Experience, DataBase Marketing, Driller Q & A, Marketing thru Operations, Measuring Engagement

Jim answers questions from fellow Drillers

Topic Overview

Hi again folks, Jim Novo here.

Sometimes the traditional RFM model does not work very well for a specific business model. For example, small business databases can be too small to fill out all 125 RFM segments properly, resuslting in distortions of predictive capability. Optimizing the traditional RFM approach for unique business model criteria is a very useful skill, and it’s actually not difficult if you understand the levers of the business model. To Drill or not to Drill, that is the question …

Q:  I’ve used your site a lot and found it to be very informative.

A: Thanks for the kind words!

Q:  I have a question about the use of RFM analysis for a low margin, eCommerce business.  I read that for a relatively small customer list (<50k) using just the “RF” of the RFM analysis would be preferred since the “M” tends to hide shifts in behavior.

A:  Well, the M tends to smooth shifts regardless of the size of your list.  In addition, if you have a small list, 125 segments is too many to be really useful, so RF at 25 segments in more intuitive.  The real issue with M or Monetary Value is up and coming, accelerating customers.  If you use total spend (M), it will “punish” them with a lower rank.  But the fact is they have more future potential because Recency is low and Frequency is ramping.  Inversely, M tends to reward customers who have spent a lot in the past with a higher rank, though they may actually be declining or defected customers.  Predicting the future is more profitable than reporting on the past, so given a choice, I would drop “M.”  This is especially true on the web, where communication costs are low and changes in behavior can be very rapid.

Q:  My question to you is, since I’m talking about a low margin business, wouldn’t “M” actually be more valuable than “F” for the analysis?  For example, if 40% of my customers are driving 70% of my sales and 100% of my profits, that says that 60% of my customer base is losing me money.  I don’t want them to be given a higher value rating because they’re placing MORE unprofitable orders than someone placing fewer but profitable orders.  You see what I’m saying???

Continue reading New RFM: Using RF or RM Instead of RFM →

New RFM: Snapshots versus Movies of Behavior

Analytical Culture, Analytics Education, DataBase Marketing, Driller Q & A, Marketing thru Operations, Measuring Engagement

Jim answers questions from fellow Drillers

Topic Overview

Hi again folks, Jim Novo here.

The standard RFM customer model is essentially a “snapshot” of likelihood to purchase (and so perhaps also profitability of campaign) at a given point in time. But what if you took these snapshots and turned them into a movie, looking at likelihood to purchase over time, and what specific inputs affected these changes? Then you’d have a LifeCycle movie / model, which amplifies the power of RFM substantially. Ready to find out more? The Drillin’, the Drillin’ …

Q:  I am in catalog circulation.  We currently use RFM to segment the file and then roll the RFM cells into more manageable segments (this is a new technique to me, I am new to this company, in my former company we mailed by RFM segment).

A:  Hmm…this sounds like a “dumbing down” approach to RFM, but hey, if it works, why not.  Sometimes this is done because the customer base is not really large enough to support 125 segments, and the differences between the segments can become unstable and less predictive unless they are aggregated.

Q:  Because we are in a niche market and we saturate it pretty well, I would like to see which customers are on the edge or falling off (the Latency stuff) and which ones we can “reward” for being the best.  I do not think the RFM analysis shows me that amount of detail.

Continue reading New RFM: Snapshots versus Movies of Behavior →

Retention and Defection Scoring in Travel

Analytical Culture, Analytics Education, Customer Experience, DataBase Marketing, Driller Q & A, Measuring Engagement

Jim answers questions from fellow Drillers

Topic Overview

Hi again folks, Jim Novo here.

What happens when your business doesn’t really fit the mold of traditional customer scoring that retail and similar businesses use? Well, you can certainly come up with a version of the tradional models that is customized for your business, as we do below for a fellow Driller’s travel agency. Wanna see how that’s done? Then, to the Drillin’ …

Q:  Just read your book and I say full marks for such a practical and sensible approach!!! Start small and grow is the way to go. 

A:  Well, thanks for the kind words!

Q:  I am a part owner of a travel agency (not been the best area to be in during pandemic).

A:  Eeeeek!

Q:  My first focus for Drilling Down is on our leisure customers.  But my head is spinning a bit with all the ideas I have from your book.  I can electronically access our: customer names etc., an ID number, when they purchased, how much the product cost, the supplier, the category (i.e. air only, cruise, tour etc.) and the final destination.  If you would be so kind as to give me a little steer in the right direction  in setting up the metrics and scores.

A:  Hmmm…  I of course don’t know your business but would think that particularly in leisure, there is a natural cyclicality caused by vacation timing, anniversary events, and such.  So in terms of timing, you use a classic Latency approach, e.g. if a customer took a trip last July they are somewhat likely to take one this July.  If they took one last July AND the July before, they are very likely to take one this July.  If they have taken a trip the last 5 July’s in a row, they are extremely likely to take one this July.  

So you can rank customers by likelihood to travel each month, and if you want, could assign them a “score” to represent this likelihood, in the case above, extremely likely = 5, very likely = 4, etc.  People who have not booked with you for a year might be a 2, not for 2 years a 1. 

Continue reading Retention and Defection Scoring in Travel →