|In this issue
Last issue, we told you about LSMs LeadGenius system
and the TARGUS Info system that can verify address
information against caller-id. The goal here is to
spend more time with customer's that can make you
money and less time with the tire-kickers. In other
words, get off the phone with unsavory leads and spend
time with qualified customers.
This issue we talk about using past call data to hone
your call center questions and more accurately predict
the value of inbound leads. I'll warn you it gets
technical, but this is how serious direct response
marketers make serious money. With the addition of our
onstaff Economist we're introducing you to a new
concept of yield management called ECONOMETRICS.
To frame the problem, we turn to our Legal client
seeking a very rare form of lung cancer caused by
exposure to asbestos. With only 2,700 new Mesothelioma
cases diagnosed each year, the goal is to create a net
(aka lead definition) that rewards stations for
broadcast when they send over a proper Valid Lead. The
problem? Too tight of a net and the stations stop
broadcasting due to low ROI. Too loose of a net and we
get a pile of expensive invalid leads and the client
To get you up to speed, let's look at what
Econometrics literally means 'economic measurement'.
It is a combination of mathematical economics,
statistics, economic statistics and economic theory.
The two main purposes of econometrics are to give real
world solutions from economic theory and also to
empirically verify these real world practices.
For example, one of the two most predictive tools we
found in the Asbestos project was whether a caller had
been diagnosed with Asbestos and that they smoked.
Smoking aggravates asbestos exposure into
Mesothelioma. The only problem is that Asbestosis
alone has no value to the attorney. Nor does lung
cancer alone. Folks were properly responding to the TV
commercials, answering accurately they had been
exposed to asbestos, and that they smoked. Problem
was, they did not have Mesothelioma--nor would they.
How do you wrangle in these answers, sift through
them, and get valid cases at a predictive value?
Arguably the most important tool of econometrics is
regression analysis. This means assigning a
proportionate value for each variable or question.
Say, for example, that having the condition Asbestosis
predicts 20% of the outcome for eventually getting
Mesothelioma. But it's not causal. You have many other
factors that predict the case value. So we go deeper.
Econometric analysis can often be divided into
time-series analysis and cross-sectional analysis.
Time-series analysis examines variables over time,
such as matching past response against successful case
acquisition. Cross-sectional analysis studies
relationship between different variables at a point in
time. For instance, the relationship between
geography, time-since-exposure or how many packs of
smokes a day the patient had and over what time.
When time-series analysis and cross-sectional analysis
are conducted simultaneously on the same sample, it is
called panel analysis. If the sample is different each
time, it is called pooled cross section data.
Multi-dimensional panel data analysis is conducted on
data sets that have more than two dimensions. For
example, some forecast data sets provide forecasts for
multiple target periods, conducted by multiple
forecasters, and made at multiple horizons. The three
dimensions provide more information than can be
gleaned from two dimensional panel data sets.
Econometric analysis may also be classified on the
basis of the number of relationships modelled. Single
equation methods rely on the assumption of a causal
relationship between the variable of interest (the
dependent variable) and the explanatory or exogenous
variables. If this assumption is not satisfied, the
results may be subject to simultaneous equations bias.
A variety of simultaneous equation methods have been
developed to take account of the fact that variables
such as time-since-diagnosis, age or industry you
worked within are, in general, jointly determined in
Here's what it might start to look like:
Y� = â0 + â1( Phone�) +
â2(Smoke�) + â3(Diag�)
In English, this means our new Econometrics program
developed multiple variables and are providing the
client with a multi-variable equation that weights the
callers answers in real-time.
Now instead of having a fixed lead definition, the
call center will provide a final score for client and
broadcaster that scientifically pays according to
value. Suddenly, the client can scale their $500,000
media investment into a predictable return of
$20,000,000. Plus, broadcasters can be credited fairly
on a PI basis.
Our 20 year direct response industry veteran says he
can do the same with logic--despite having a MBA from
Berkeley. It's painful for many to be this exact.
Sure, you could guess about this process and even try
to determine logically where more cases lie. But when
you're scaling media from $600K to $20 million
annually, wouldn't you like to be sure?
Getting scientific direct marketing help is easy now.
If you'd like a leg up on the competition using
Econometrics, learn more about this new media
management program from Last Second Media:
* Complete Econometrics projects available from