Drilling Down Newsletter - August 2001 - Latency
Metrics, Google AdWords
In this issue:
# Best of the Best Customer
# Tracking the Customer LifeCycle:
# Practice What You Preach:
Online Advertising Effectiveness?
Tell Me About It... (Part 4)
# Questions from Fellow Drillers - suspended
for this issue, see below
Hi again folks, Jim Novo here. This month we've got a little
change in format. We'll start with some hot links as usual.
But I was inundated with questions on the Latency piece from last
month, so I'm answering these questions as part of the series on
Tracking the Customer LifeCycle, and skipping the
"Questions" section for this month. Plus, we finally
get some resolution on the Advertising Effectiveness question...or did
Let's do some Drillin'!
Best of the Best Customer Retention Articles
This section usually has links to "must read" articles about
to move into the pay-to-read archive at DM News. I'm afraid the
content has been a little less than "must read" from them
lately. So here's a few important articles from elsewhere on the
web in case you missed them, all having a very specialized focus.
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
every 2 weeks before they expire by subscribing to the newsletter.
- Loyalty Builder or Loyalty Loser
July 18, 2001 Colloquy
This article may require free registration to access; if you care
about loyalty programs, you should already be registered at this site.
A critical, reasoned look at why "point exchange" or
networked loyalty programs will not deliver the results you are
looking for in a loyalty program for your business.
Customers May be Different Tomorrow
July 19, 2001 crm.ittoolbox.com
This is an excellent look at customer analysis in the banking
industry. Forget the 80/20 rule, how about 150/20, which means a
huge amount of customers have negative value. Great stats in
this article - if you are into bank stats that is.
Without Spending $1000/M
July 26, 2001 Catalog Success
Web sites have had good results using targeted offline media like
catalogs and direct mail to generate visits and sales. Next
step? Targeted space buys. This was written for
catalogers, but it does apply to onliners - trust me. Lots of
tips on buying targeted space.
Tracking the Customer LifeCycle:
Last month, I introduced the Customer LifeCycle metric of Latency, or
the amount of time passing between two customer actions. This
time passed is usually measured in weeks or months offline, and in
days or weeks online. If you would like to review this article,
here is a link to it:
Customer LifeCycle: Latency
No question about it, the constant drumbeat of the CRM machine over
the past several years has confused the heck out of people. I've been
doing this stuff for almost 20 years now, and I can tell you it is not
as difficult as it is often portrayed. Sure, you can make it
very, very complicated if you want to. But if you don't start
with the basics, you're going to end up wasting a ton of money.
Let's start simple, shall we?
I'm going to back up a second and explain in a more general sense how
metrics like Latency are used, and in particular, address some of the
misconceptions people have regarding customer value-based and
relationship marketing techniques. Much of CRM is based on these
fundamental ideas. Remember, CRM is an approach to managing a
business, not a technology. You do not need to live on the
bleeding edge of technology to take advantage of a customer-based
Generally, CRM or Relationship Marketing attempts to define
customer behavior and then looks for variances in behavior. When
you hear people talk about "predictive modeling" or looking
for "patterns" using data mining, they are essentially
taking a behavioral approach using the latest tools. Once you
know how "normal" customers behave, you can do two things
with your business:
* Formally document normal customer behavior and internalize
it systemically, leveraging what you know to improve business
functionality and profitability.
* Set up early warning systems, triggering events, or "trip
wires" to alert you to customer behavior outside the norm. This
variance in behavior generally signals an opportunity to take action
with the customer and increase their value - inline or offline.
Customers in the aggregate tend to follow similar behavioral
patterns, and when any single customer deviates from the norm, this
can be a sign of trouble (or opportunity) ahead. For example, if
the average new cellular customer calls customer service 60 days after
they start, and an individual customer calls customer service 5 days
after they start, this customer is exhibiting behavior far outside the
norm. Is there a potential problem, or opportunity? Is the
customer having difficulty understanding how to use advanced services
on the phone, or is the customer happily inquiring about adding on
more services? In either case, there is an opportunity to
increase the value of the customer, if you have the ability to
recognize the opportunity and react to it in a timely way.
Understand, there is no "average customer," and a
business will have many different customer groups, each exhibiting
their own kind of "normal" behavior. The tools
available to identify and differentiate customer segments using
behavioral metrics are discussed at length on almost every page of
this web site. For example, the type of media or offer used to
attract the customer can have a dramatic effect on long term behavior,
and customers who come into the business on the same media and offer
will tend to behave in similar ways over time.
In the cell phone case above, the measurement of Latency (number of
days until customer service call) serves as the "trip wire,"
a raising of the hand by the customer, to say to the marketer
"I'm different. Pay attention to me." It is then
up to the marketing behaviorist to determine the next course of
action. Metrics like Latency provide the framework for setting
up the capability to recognize the opportunity for increasing customer
This raising of the hand by customers, and the reaction by
marketers, is the feedback loop at the center of Relationship or
LifeCycle based marketing. It's a repeating Action - Reaction -
Feedback cycle. The customer raises the hand, the marketer
Reacts. The customer provides Feedback through Action - perhaps
they cancel service, or perhaps they add service. The marketer
reacts to this Action, perhaps with a win-back campaign, or with a
thank you note. It's a constant (and mostly non-verbal)
conversation, an ongoing relationship with the customer which requires
interaction to sustain. It is not a relationship in the
"buddy-buddy" sense. Customers don't want to be
friends with a company, they want the company to be responsive to
their needs - even if they never come out and state them openly to
This relationship continues to cycle over and over as long as there
is value in the relationship for both the customer and the marketer.
If the customer takes an Action and there is no Reaction from the
marketer, value begins to disappear for the customer, and they may
defect. When value disappears for the marketer (the customer
stops taking Action / providing Feedback), marketers should stop
spending incremental money on the customer.
Notice I did not say "fire the customer" or any of the
related drivel thrown around in some of the CRM venues. All
customers deserve (and pay for) a certain level of support. The
real question is this: for each incremental, or additional dollar
spent on marketing to the customer, is there a Return On the
Investment? If I have the ability to choose between spending $1
on a customer returning $1.10, and $1 on another customer returning
$3, I would be nuts not to choose the customer returning $3. I
have not "fired" the customer returning only $1.10; I have
just chosen not to spend incremental money doing any special marketing
Do you see the difference?
In fact, much of the profitability typical of high ROI Customer
Marketing techniques comes from knowing who not to spend on.
Most of the decreased profitability in any marketing program is a
result of over-spending on unsuitable targets with lowered returns.
But because marketers tend to look at results in the aggregate, or
they are looking at demographically-based segments to measure a
behaviorally-based outcome, they miss details like certain segments
returned $5 for each $1 spent, and others lost $5 for every $1 spent.
The campaign as a whole may return only $1.10 for each dollar spent
because the marketer spent money on low Return On Investment
When you are trying to encourage a customer to buy something, you
are looking for a behavior to occur. To measure the results of
such a marketing campaign using only demographic segmentation without
any behavior-based metrics (like Recency or Latency) is misleading at
best, and lazy otherwise - it's apples and oranges.
Why is this all important?
Customers who are in the process of changing their behavior -
either accelerating their relationship with you, or terminating their
relationship with you - are the highest potential return customers
from a marketing perspective. They represent the opportunity to
use leverage, to make the highest possible impact with your marketing
dollar. You may make money marketing to customers who are just
cruising along the LifeCycle, acting like an "average
But when you can predict the likelihood of an average customer to
turn into a best customer, and you successfully encourage this
behavior, or you can reverse a customer defection before it happens,
there are tremendously profitable longer-term implications for the
bottom line. You discover these opportunities by understanding
behavior and setting up trip wires (like Latency metrics) to alert you
to deviations from normal behavior by a customer.
What about all the rest of the customers, those who are not either
accelerating or terminating the relationship? Leave 'em alone.
Whatever background marketing you do (advertising, branding, service
campaigns, etc.) is serving them just fine.
High ROI data-driven marketing techniques are best used (and create
the highest returns) when they are used to surgically strike at a
trend in behavior, not when customers are comfortably plodding along.
However, there's not as many comfortable plodders as you think;
in fact, from 40% to 60% of your customer base is either in the
process of accelerating or terminating their relationship with you
right now. The question is, how do you take advantage of the
Latency, and all the other metrics described on the Drilling Down
site, are simply tools for recognizing the opportunity to take an
Action in Reaction to the customer raising their hand. If you
don't have some kind of system to recognize customers in the process
of changing their behavior, you will miss out on most of the highest
ROI customer marketing opportunities you have.
And don't count on the customer to e-mail you when they're thinking
of changing their behavior - we both know that just is not going to
happen. A more likely scenario: they will just stop taking
Action and providing Feedback. And by then, it's too late for
you to do anything profitable about it.
Set up your trip wires and predict the behavior, folks. It's
the only way to sense when an average customer is ready to become a
best customer. And reacting to a customer defection after the
fact is a truly sub-optimal way to "manage" a
If you'd like to see more on LifeCycle-based marketing in future
newsletters, be sure and let me
Practice What You Preach: Online Advertising
Effectiveness? Tell Me About It... (Part 4)
Last month we took a look at the quality of visitors generated by my
paid search listing ads on Google and GoTo. The following table
compares visitors using the 3 most popular search terms to find my
site with the "average" paid search visitor. RM =
Relationship Marketing, CR = Customer Retention, CL = Customer
Loyalty, and TS = Total Site statistics, all paid search listings:
Avg. Visit Length 8.49 8.44
% 1 Page Visits 24%
22% 20% 43%
% Downloading 8.2% 6.1%
% Bookmarking 9.6% 7.6% 12.2%
% Subscribing 4.5%
4.5% 2.4% 3.2%
Well, we're getting there. We've previously proved visitors
clicking on a paid listing are of higher quality than "free
search" visitors for the same search term, and now we see there
is also significant variability in quality of visitor by the term
itself, according to the chart above. Look at Customer Loyalty
(CL). Much shorter visits, and lower download and newsletter
subscribe percentages, but much higher bookmarking percentages. What
could this difference in behavior mean?
The stats above are a combination of all visitors for the same
terms from both Google AdWords and GoTo, so it seems logical the next
"Drill Down" would be to look at each source individually,
and that is just what I have done. For clarity, instead of
creating two charts and having you bust your eyeballs trying to
compare them, I have created a *ratio* between the Google and GoTo
Google to GoTo Ratio
Avg. Visit Length 65%
% 1 Page Visits 110%
48% 112% 570%
% Bookmarking 72%
If you were to read down the Relationship Marketing (RM) column, this
"For the paid search term Relationship Marketing, the Average
Visit Length for visitors from Google is 65% that of GoTo, the percent
one page visits is 110% that of GoTo, the percent Downloading is 48%
that of GoTo," and so on. A number over 100% means Google
is higher than GoTo, under 100% means Google is lower than GoTo.
One thing is perfectly clear from this chart - Google dramatically
under-performs GoTo for the paid search term Relationship Marketing,
and outperforms GoTo on the paid search term Customer Loyalty, across
the board, in every category (note a lower number on
% 1 Page Visits is better).
Things are less clear-cut for the term Customer Retention, although
I'd have to give it to GoTo because Bookmarking and Subscribing to the
newsletter are highly correlated to future purchase of a book.
Where does this leave us? Overall, it appears you can not
attribute "quality" as defined here to either a search term
or a search engine; there is a * combined * contribution which creates
dramatic visitor quality differences. This is a perfect example
of the mistake people make when using "averages" or looking
at the "average customer" - rarely does the average customer
represent the true underlying behavior of the actual customers.
Tactically, it means I should budget paid search expenses by term
by engine, and in the case above, shift most if not all the budget for
Relationship Marketing to GoTo, and most if not all the budget for
Customer Loyalty to Google. Customer Retention might need a
little more work to resolve, but instead of running the budget 50 / 50
as initially set up, it would make sense to maybe run 70% on GoTo, and
30% on Google, from what I see here. Hey, it doesn't always come out
black and white, you know?
As far as why this occurs, it's fun to speculate, but a
marketing behaviorist cares more that it does happen -
it's a fact, Jack - and takes action based on this fact. There's
plenty of time to wonder about it later, after the spending has been
reallocated and the highest ROI possible is being realized.
A "gun to the head" guess? It's the content at the
other end of the click making the difference. The content on the
Customer Loyalty page appeals more to a Google user, and the content
on the Relationship Marketing page appeals more to a GoTo user.
Why? I haven't got a clue. Check them out for yourself:
Loyalty (favored by Google user)
Marketing (favored by GoTo user)
Let me know what you think. If the responses seem to be
trending one way or the other, I'll present the arguments in the next
newsletter. Meanwhile, the idea of content making the difference
(a 3rd variable in addition to term and engine?) is kind of
interesting - maybe there's a way to test the idea.
I'll let you know...
If you'd like to see more on web log analysis
in future newsletters or comment on the Google versus GoTo page
preference issue, contact me using
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 to me, 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 & when the material will appear.