Intra-Company Promotional Risks

Jim answers questions from fellow Drillers
(More questions with answers here, Work Overview here, Index of concepts here)

Topic Overview

Hi again folks, Jim Novo here.

Eating Your Own is not a great idea. Yet in many large companies, different divisions literally try to steal customers / sales from each other using the common customer database. Sure, everyone gets real excited over the plans to merge databases across the company and get the “full view” of the customer, and it makes perfect sense. Perfect sense – if you’re also institutionally prepared for wild battle over customer ownership and value stealing to come. This is particularly true (and difficult to deal with) in multi-channel retail environments.

Got your Drillin’ shows on? Let’s take a stroll through subsidy cost land …


Q:  Hi Jim,

Our company has 8 divisions and we completed the integration of all the customer databases a couple of years ago.  We have the 360 degree view of the customer, at least as far as sales transactions, across the entire company in one database.

A:  Congratulations!  However, I note not joy, but some kind of concern in you voice.  I’m just waiting for the “But…”

Q:  The database services group I am part of is under IT.  We respond to requests from the different divisions for customer analysis and the creation of promotional lists for email, direct mail, and telephone campaigns.  It’s interesting because our group is finally directly involved with increasing the profitability of the business and we have some input, which makes the job more rewarding.  I picked up your book because I thought reading it might increase our ability to contribute.

A:  Well, again, congratulations!  But I’m still getting that nagging feeling from your tone.  Still waiting for the “But…”

Q:  I’ve got a two part question for you:

1.  What I am seeing is the different divisions promote to “best customers” of the company as a whole, or even try to target best customers of another division for their campaigns.  It seems to me that contacting these same people over and over from the different perspectives of the divisions is not optimizing customer value, and might actually be irritating to the customers (I know it would be irritating to me).  

The contact frequency across all divisions to the same customer can reach 4 – 6 times a month through various media (phone, mail, e-mail).  Also, there is no customer retention effort going on that I am aware of, it appears it is all acquisition oriented but the main targets are customers of other divisions.

A:  Oh boy, the database marketing pendulum has swung the other way, from “we don’t know what to do with all this data” to “we’re really maximizing that database asset”. You are correct to be concerned about this issue.  Talk about “push marketing”… the intent of each division is “pull” because of the targeting but the result is “push” because of the volume and “noise” created at the customer level.  Too much too fast.

Continue reading Intra-Company Promotional Risks

Discovering Customer LifeCycles

Jim answers questions from fellow Drillers
(More questions with answers here, Work Overview here, Index of concepts here)

Topic Overview

Hi again folks, Jim Novo here.

Today, we’re asked for a simple definition of retention. Problem is, the data / biz model really creates the definition. Meaning, you gotta match the creation of metrics with the actual actions.  So I call for segmentation first so we can put some “actionable” stuff in the mix.

Make sense? Let’s do the “simple” (easy? maybe not) Drillin’ …


Q:  For an online retailer, what is the best way to gauge retention in its most basic and simplest form?  % of orders that are from repeat buyers?  % of orders in month 2 who are repeaters that first bought in month 1?

A: I would take direction on this from the actual results of campaigns.  Basically, at the point a customer no longer responds, they have defected.  Perhaps this averages 3 months or 6 months after 1st purchase, and there will be category or price segments within these “time” segments.  Retention is really measured by the defection.

Now, that’s not to say that % orders from repeats or the other one you mentioned are not valid, but I suggest you think about the specific  question you want answered by the metric you choose.  % orders from repeats, for example, is a common metric in mail order but is often biased by campaigns, e.g. if you ratchet up customer acquisition during a single month you poison your own metrics.

Continue reading Discovering Customer LifeCycles

Modeling Customer Behavior with Small Databases

Jim answers questions from fellow Drillers
(More questions with answers here, Work Overview here, Index of concepts here)

Topic Overview

Hi again folks, Jim Novo here.

We’re about to take a trip into the world of small scale databases. In particular, how does a not-for-profit with a small database of donors go about using predictive models? Answer: Keep it simple. Try to avoid using a lot of variables; look for the most powerful and stick with those until you are able to uncover additonal info and grow the database. Ready?


Q: I am new in the NFP (Jim’s note – Not For Profit) sector and would like some advice re:  segmentation models to optimize campaign results – both response and value (Short Term  and Long Term).  Do you know or is there any knowledgebase of how the various techniques – behavioural, RFM, demographic, geographic – generally rate against each other?

A: Not other than my web site / book, which generally covers all the simple models. There is plenty of info around on the web though.

Assuming the end Objective is a donation, the behavioral stuff is going to be much more productive than the geo / demographics are. It’s like a pyramid.  My friend Avinash “stole” (with my permission) a slide from my presentation on this topic and put it on the web, you can see it here.  You’re looking for an “action” (donation), so actions (behavior) will be the most useful segmentation, at least as a primary cut.  Then you can get into geo /  demo stuff if it improves the model.

Q: As my database is small I don’t have the luxury of testing multiple techniques and causal factors.  I will probably run tests in series but would like a general idea of which ones to test first to cut down the time.

A:  Not sure what you mean by “small”, but in general, the more complex a behavioral segmentation approach is the larger the database it needs to be useful.  So for example, with classic RFM (125 segment scores), the bare minimum for it to make any sense is probably 5000 records, and you should really have at least 10,000.

Continue reading Modeling Customer Behavior with Small Databases