|
Modeling
at Mal Dunn Associates, Inc.
Mal Dunn Associates, Inc. performs marketing magic by providing
our clients advanced marketing technologies in data-rich,
sophisticated database environments at an affordable costs.
Mal Dunn Associates takes a unique approach to modeling in
that we are not tied to specific statistical techniques. Mal
Dunn Associates Modeling uses a comprehensive analytic strategy
designed to refine the marketing program that will meet your
business goals, utilizing whatever techniques necessary to
identify your best prospects. A few of the techniques are
noted below. In all cases, Mal Dunn Associates’s analysis
is determined by your marketing needs- not the other way around.
----------------------------------------------------------------------------------------------------------------------
Prospect File Modeling
Good Customer Match
The mailer supplies us with a sample of their “best
customers”. Mal Dunn Associates will create a good customer
model identifying the buyers, subscribers and donors the target
file that most closely resemble the mailer’s best customers.
Mailed
Model Regression
A test of a segment is required. The responses to that mailing
are then modeled and the database is then scored based on
this response model.Generally speaking, response models have
had even more success than good customer matches. This is
because these models are based on responses to a specific
mailer and a specific mail piece. Both responder data and
non-responder data is used to build the model.
Previously
Mailed Match Model
This approach combines the speed and cost of a Good Customer
Match with the effectiveness of a Response Model. The end
users last mailing (which has the responses keyed) is matched
up against the database to be modeled. A response model is
then built as if the database was just mailed. This is a
way to use the mailer’s specific response data without having
to mail a sample of the file to be modeled. Costs are similar
to those of a Good Customer Match. We recommend this for mailers
who have had little success with Good Customer Matches in
the past but don’t have the time to mail a sample of
the file for the initial test.
Zip-Level
Models
This is similar to the Previously Mailed Matched Model except
that ZIP codes are ranked from most to least responsive. This
model is applicable to all lists and creates an inexpensive
and fast way to further target response.
Shadow
File Models
Mailers who have had success renting names from MarketShare
often use this option. Here, our analytic staff will take
an existing segment of MarketShare that a mailer is successfully
using and model the rest of the database to identify the buyers
and subscribers who most closely resemble the names that make
up the working segment. This is an effective way to increase
the quantity in a proven universe. Pricing and turn around
time are similar to Good Customer Matches. Geo- Based Models,
Hybrid Models and Customized Reactivation Models are also
available upon request.
----------------------------------------------------------------------------------------------------------------------
House File Modeling
Customer Reactivation
This method develops response models form previous expired
mailings. It can identify recent buyers of other products
or compare your best customers to your expires allowing you
to profitably turn ex-customers into new customers again.
House File Analysis
Using your internal data and enhancing it with external RFM/source,
demographic and lifestyle data, will enable you to determine
which offer will maximize your response.
|
