Drilling Down Newsletter - May 2001 -
Overture Optimization, Recency Promo

In this issue:

#  Best of the Best Customer Retention Links
#  New Recency-based Tutorial on Web Site
# Optimizing Discounts Using Recency
# Practice What You Preach:
   Online Advertising Effectiveness?  
   Tell Me About It...
# Questions from Fellow Drillers
------------------------------------
Hi again folks, Jim Novo here.  This month we've got great customer retention article links, and a new tutorial on the web site.  We're going to take a look at discount "laddering" using Recency, and take a brief run at some puzzling WebTrends stats on search engine traffic.  And, we'll run through two quick questions from subscribers to the newsletter.  

Let's do some Drillin'!

Customer Retention Links
====================
No "must read" expiring articles on the DM News site this month; here are 3 great customer marketing articles you may have missed from other sources.  

Note to web site visitors: These links may have expired by the time you read this.  You can get these "must read" links e-mailed to you each month 2 weeks before they expire by subscribing to the newsletter.

Keeping Track: Is the Net Too Measurable?
May 1, 2001   eMarketing Magazine
Best article in a while on customer analysis.  This article is a good overview of what's going on, including comments from some of our peers who are now using a few of the techniques described on the Drilling Down site.  I love it when they talk Recency and Frequency (I need to get out more).

On Average
May 2, 2001  DM Review
A very simple explanation of what the 80/20 rule really means.  As I've said many times before, there is no such thing as an average customer, and in many cases, using this metric will just get you into trouble.  This article covers simple decile analysis, the beginnings of the work you should always do Pre-CRM; it will save you time and money, guaranteed.

Thinking Small
May 2, 2001   Darwin Magazine
More from Darwin - I think I'm going to like this mag.  Great story on Best Buy, and about KISS-ing CRM, which I keep trying to hammer home on the site.  These guys are not doing anything very fancy, just the nuts & bolts of tracking behavior and looking for "trigger points" - the subject of my book.

New Recency-based Tutorial on the Web Site
=====================
Well, I've taken your advice (thanks for the input, by the way) and put up a tutorial that goes through a simple real-life example of how to use one of the Drilling Down techniques.  The tutorial, called Comparing the Potential Value of Customer Groups, demonstrates one of the 3 key concepts vital to a successful customer retention program - attaching future value to a customer based on their source and behavior.  The example used is Comparing the Potential Value of new customers coming from 2 different ads, and follows the customers all the way through the LifeCycle, LifeTime Value, and ROI issues.  I think you'll dig it; let me know if you do!  See:
http://www.jimnovo.com/Recency-Model.htm

What are the 2 other key issues?  Managing offers for maximum response at lowest cost and setting up an "early warning system" to flag potential customer defections.  I'll talk a little about managing offers next, and as for setting up an early warning system to flag potential customer defections, well,  you'll have to get the book for that!  See:
http://www.booklocker.com/jimnovo 

Optimizing Discounts Using Recency
=====================
For those who don't know what "Recency" is all about (are there any left?), allow me a short introduction before we get to the managing of discounts using Recency.

You will generally see response rates fall as a function of Recency - the number of weeks or months since the last interaction with the customer.  This relationship is a very  smooth curve and quite predictable once you establish the "slope" of it for your business.  The response rate by Recency will look something like this:

Time since contact = 1 month, 
Response rate =  20%

Time since contact = 2 months, 
Response rate = 10%

Time since contact = 3 months, 
Response rate =  4%

Time since contact = 4 months, 
Response rate =  1%

Again, the "absolute" response rates will be different depending on the business, media used, and offer, but the "relative" response rates will follow a decelerating curve as shown above, that is, the less Recent the customer, the more dramatic a drop in response rate you will get.

In terms of using this information for promotions, you will find some point along the curve where you will get "breakeven," meaning the cost of the campaign will  equal the benefit generated.  For example, let's say you offer a discount or gift in your retention / lapsed customer campaign and need a response rate of at least 4% to pay back the campaign cost.

The implication for your campaign in the Recency information above is this: don't bother to promote to any customer who hasn't made contact in over 3 months, because you're wasting your money; response will be too low to pay back the cost of the campaign in the "over 3 month Recency" group of customers.

This Recency effect is very stable over time, allowing you to predict in advance what response to a campaign will be, once you do an "establishing" campaign to see what your response rate curve is for any particular offer at each Recency level.  With me so far?  Recency will predict response rate.

What many people don't know is if you "ladder" discounts according to Recency, you will boost overall response while cutting costs.  Let's say you usually  e-mail everybody a 10% discount.  If you were using a ladder approach for a test, it might look like this:

Time elapsed = 1 month, 
Response rate =  20%, discount = 5%

Time elapsed = 2 months, 
Response rate = 10%, discount = 10%

Time elapsed = 3 months, 
Response rate =  4%, discount = 15%

Time elapsed = 4 months, 
Response rate =  1%, discount = 20%

Using this approach, you are allocating the most "bang for the buck" discount-wise where you need it most - the least Recent customers, and pulling back on some discounting where you don't need it as much - the most Recent customers. 

Now, as I said above, your response rates will vary depending on the offer and your business, and you have to test these ladders to find the optimum profitability for each level.  The interesting and quite useful benefit of this approach is the "automatic" overall customer retention effect discount ladders have.

Using a ladder of this type means your promotional budget is automatically working harder and harder to keep a customer active with you as they drift further and further away from you.  The less Recent a customer is, the less likely they are to buy or visit again, and by using a discount ladder you are counteracting this LifeCycle process with stronger discounts as the defecting customer behavior plays out.

If a most Recent customer does not respond to the 5% offer, as they get less Recent, they "automatically" get offers rising in value, and at some point, many will take it.  The customers who run through this system without taking any offers were likely lost to you as a customer already, and not worth the extra expense to try and keep promoting to them.  Clean, simple, easy to implement.  And if you don't have any formal customer retention program, much better than what you're using!

Practice What You Preach:
Online Advertising Effectiveness?  
Tell Me About It.  (Webtrends)
=====================
Well, I got some positive feedback on the last Webtrends article so I figured I would toss in another.  I don't want to sound like a shill for WebTrends, but I don't know how you manage a web business without detailed log  analysis.  WebTrends is not nearly as good as the system I used for the CBS/SportsLine "points for page views" loyalty program, but then again, not many of you probably need something with that much horsepower.  Or do you?  Let me know...

Take my current little pet peeve - I'm getting ripped off on advertising, it would seem.  Or am I?  Oh, not on the response rate side, I get great response rates with Google AdWords and GoTo.  I'm talking about the quality of the visitors generated.  It seems that visitors coming from my ads might be of a lower quality than  free visitors coming directly through the search engines.

Check out this little chart, all based on visitor sessions:

Metric                 |Ad Visitors |Search Visitors
________________________________
Avg.  Visit Length      7:43             9:12
% 1 Page Visits            38%             51%
% >10 Page Visits         9%                5%
% > 19 minute Visits   11%                7%

% Downloading Book Sample   
                                    2.2%           10.2%
% Bookmarking Site    
                                    2.2%            11.0%
% Newsletter Subscribes
                                    3.7%             5.1%

Hmmm, he said.  Not much to make a decision on here, but the differences are striking enough to warrant further investigation I'd say.  The page viewing activity seems to indicate the ad-driven visitors are of higher quality (lower one page visits, higher percentage of high activity users) but the "engagement behavior" of the search-driven visitors (downloading, bookmarking, subscribing) is far more valuable, as these visitors are most likely to turn into book buyers.  What's really going on here?  We'll "drill down" another level next month and try to find out.
--------------------------------
If you'd like to see more on web log analysis
in future newsletters,  let me know.
-------------------------------

Questions from Fellow Drillers
=====================
Hi Jim,

Q: I have a large client with an opt-in e-mail campaign.  We found that the majority of customers who subscribe to these e-mails have never clicked on the offers they contain, and about one third have never even opened  an e-mail.  Can you direct me to any industry best practices/tactics of dealing with these inactive populations? 

A: Well, I don't know about any "industry best practices" on this topic for e-mail, but I can tell you what is done offline - threaten 'em and then purge 'em.  There is evidence for this working online, as it has been done successfully by people online wanting to maintain high list quality and responsiveness (the same reasons it is done offline). 

There also seems to be some evidence that a "hey, if you're not going to read the e-mail, we're going to stop sending it" threat in the subject line, when applied with a "click here to confirm you want to keep getting this e-mail" in the body, actually converts people who were previously inactive to active openers, readers, and responders. 

Again, this is what is seen offline.  That's why you get catalogs with the similar "This is your last catalog if you don't buy" threat.  It's a winner either way for the marketer; you make some additional sales from dormant targets and you stop spending effort on the ones who really don't care about your offers in the first place.  More revenue, less cost.  The Drilling Down mantra, for sure.

Q: You seem to provide practical methods for turning data into useful customer information.   Do you think the lack of this expertise is the biggest reason why data-driven marketing is not more widely used?

A: Well, I don't know about not widely used.  The $110 billion catalog industry is built on data-driven marketing.  There are some people on the web doing a good job with it, though most of those are catalog operations.  But yes, in general people are confused about which data is important and how to use it.  There just are not many people around with practical hands-on experience.  Dealing with data is also highly math-oriented, and a lot of people don't feel comfortable with running numbers.  These are the reasons I wrote the book.  Techniques like mine are explained in some books, but either at too high of a level to be of practical use or with such mind-numbing detail and math you go crazy reading it.  One day I thought to myself,  Hey, you can strip this all down into a very basic approach, and really show the average person how to make a lot more money in customer marketing - by giving them the "dummies version" of advanced database marketing techniques.  And to get rid of a lot of the math, I use graphical displays of behavior so "getting the results" is much easier.  In a nutshell, that's the book.
------------------------------

That's it for this month's edition of the Drilling Down newsletter.  If you like the newsletter, please forward it to a friend!  Subscription instructions are at the top and bottom.

Any comments on the newsletter (it's too long, too short, topic suggestions, etc.) please send them right along, with any other questions on customer Valuation, Retention, Loyalty, and Defection to me.

'Til next time, keep Drilling Down!

Jim Novo

Copyright 2001, The Drilling Down Project by Jim Novo.  All rights reserved.  You are free to use material from this newsletter in whole or in part as long as you include complete attribution, including live web site link and/or e-mail link.  Please tell me where and when the reference will appear.