1

Using External Databases

DIRECT MARKETING CHAPTER 4
Ass.Prof.Dr. Tuğçe Ozansoy Çadırcı
 2

USING EXTERNAL DATABASES

External
Databases

PART 1
Using External Databases
3
The Best Data You Will Ever Get

‣… is that on your own database: it relates to your

customers and their transactions with your
company.
‣If you are expanding your business – or have just
identified a new market or product – the data you
have may need augmenting.
‣This is where external databases come in.
Using External Databases
4
Direct Marketers Use External Databases For Two Reasons

‣To add data to the existing database in order to

understand their customer better.
‣When they decide to enter new markets:
‣To get accurate profiles of new markets.
‣To decide who to contact with.
Using External Databases
5
Main Areas in External Databases

‣Geo-demographic Products:
‣Households are segmented according to demographic
characteristics.
‣This segmentation is added to postcodes and mailing lists can
be created.
‣Lifestyle Databases
‣The answers received to questionnaires, product registration
forms.
‣ Attitudinal data
Using External Databases
6
Profiling Using External Databases

‣Step 1. A “matching process” is undertaken between

the internal and external databases
‣Matching: the computer identifies the common names, post
codes held in both databases.
‣ Geo-demographic: post codes are matched.
‣ Lifestyle: names and addresses can be matched.
‣Step 2. The success and effectiveness of the
matching process highly depends on whose
database the company uses, and how well the
company’s database is structured.
‣If the company is using a geo-demographic database the post
codes should be matched 100%.
Using External Databases
7
Profiling Using External Databases - 2

‣Step 3. The matched and combined data is used to

form customer clusters/segments. These clusters
should be relevant to the company’s products.
‣Step 4. Clustering for a new product/market.
‣The generated prospect list is ranked in order of greatest
similarity to the existing customers.
 Using External Databases
8
Profiling Using External Databases - 3
Firm’s customer le + geo-demographic or lifestyle database
Look for direct matches of names and addresses

From one million customer les, say 200.000 direct matches are found

Matches are fully pro led; i.e. described quan ta vely by descrip ve modeling techniques:
What are the characteris cs which cluster them together?

The choice now is to focus on the client’s internal database (1) or the geo-demographic/lifestyle operator’s
prospects database (2)

(1) (2)
The pro led match is compared with the internal The pro led match is compared with the
database remainder of the external database

Products likely prospects, scored by their

Produces segmenta on and research possibili es
similarity to pro le

Figure'4.1!!Using'profiling'to'iden4fy'prospects'on'an'external'database'
fi
fi
fi
fi
ti
fi
ti
fi
ti
ti
ti
ti
Using External Databases
9
The Matching Process
Client’s)
database)
Prospect)list)
for)client)
Matched)
customers)
profiled)

Prospects)
Iden0fied) External)
database)

Figure)4.2!!Using)a)matching)process)to)profile)customers)
Using External Databases
10
Using External Databases for Market Research/Segmentation

‣No attitudinal data

‣Common approach is conducting a marketing research
‣When external databases are used these problems are solved
naturally.
‣Geo-demographic and lifestyle data provides an understanding
of why people buy.
‣External data enhances the quality and extent of profile data
Using External Databases
11
Lifestyle surveys vs. Conventional Research

Conven&onal)research Research)based)on)lifestyle)surveys
High)cost)–)Even%at%the%minimum%level%of%sta0s0cal% Low)cost)–)Design,%prin0ng%and%distribu0on%of%
significance.%% ques0onnaires,%the%data%capture%and%the%analysis%of%
Very%expensive%at%any%substan0al%scale.%% results%should%give%a%cost%of%just%pence%per%
Costs%closely%related%to%number%of%persons%surveyed. respondent.%

Use)restricted)to)research)applicants)–)Sta0s0cal% Respondents)can)be)mailed)–)Produces%a%poten0ally%
tabula0ons%are%the%sole%output:%MRS%rules%posi0vely% valuable%list%of%selfGsubmiEed%names%and%addresses,%
forbid%the%aEribu0on%of%names%to%research%data.%% with%a%wide%range%of%informa0on%about%each.%%

Inflexible)–)20/20%vision%is%required%at%the%0me%of% Ease)of)followBup)–)The%list%of%names%is%available,%any%
designing%the%ques0onnaire.%Not%prac0cable%to%look% kind%of%followGup%suggested%by%the%results%of%the%
at%the%response%and%discover%some%interes0ng% original%ques0onnaire%is%simple.%%
further%ques0on%one%would%like%to%explore%with%
certain%respondents.%%

Table&4.1!!Conven.onal&research&vs&research&based&on&lifestyle&surveys&
Source:&Webber,&2002,&pp.&3.5–18.&
Using External Databases
12
Advantages of Market Research

‣Accuracy of data
‣Respondent anonymity
‣ Detailed understanding of the respondent attitudes and beliefs.
‣Surveys designed for a specific purpose.
‣ More accurate and in-dept data
‣Representativeness
‣Sample design should ensure this.
‣ The lifestyle databases lack of representativeness; as they don’t include the
information about the people who did not completed the survey.
Using External Databases
13
Using External Databases For Segmentation

‣Product needs are often “related to” or even “caused

by” lifestyle or demographic factors.
‣Internal data is often more useful in segmentation
processes
‣In situation where internal data needs an
enhancement external lists can be used for
segmentation purposes.
Using External Databases
14
External Databases as Lists of New Markets
General'Marke*ng' Direct'Marke*ng'
Segmenta(on*and*targe(ng*are*aided*by* Segmenta(on*established*by*understanding*of*
market*research* exis(ng*customers,*overlaying*external*data.*
* Profiles*based*on*large*volumes*of*data*
* *
Targe(ng*opera(onalized*by*media/retail* Targe(ng*achieved*by*exact*match*of*accurate*
choice* profiles*to*wellDunderstood*new*market*
* profiles*
* *
Brand*building*and*communica(on*to*a*mass* Communica(on*goes*precisely*to*profiled*
audience:*awareness*and*interest*build*up* prospects.*Can*be*segmented*according*to*
* type*of*prospect,*e.g.*compe(tor*buying*could*
* be*acknowledged**
* *
Sales*promo(on*in*store*or*in*pack:*mostly* Sales*promo(ons*directed*to*definite*
wasted*on*exis(ng*customers* prospects*only*
Table*4.2!!Comparision*of*general*marke(ng*and*direct*marke(ng*opera(ons*for*the*aHrac(on*of*
new*markets*
Using External Databases
15
Prospects Lists: Strategic Options

Existing Products New Products

Market' Product'
Existing Markets Penetra*on' Development'

New Markets Market'

Diversifica*on'
Development'

Figure'4.3!!Ansoff’s'Matrix'
 Using External Databases
16
Prospects Lists: Strategic Options
‣ Product needs are often “related to” or even “caused by” lifestyle or
demographic factors.
‣ Market penetration
‣ List of competitor customers
‣ Information about direct competitors’ customers.
‣ Direct marketers try to identify the prospects for brand switching.
‣ Market expansion
‣ Lists of prospects from the profiles of existing customers.
‣ Product Development
‣ Same customers.
‣ External lists are used as supplements of the internal data to define
profiles.
‣ Market Development
‣ Lists of prospects in segments where we are currently not well
represented
‣ Can be backed up with market research.
 17

USING EXTERNAL DATABASES

External
Databases
Explained
PART 2
Using External Databases
18
Geo-demographic Databases

‣People have a desire to associate with people similar

to themselves and they have a basic instinct to form
into groups.
‣Geo-demographic profiling suggests that there are a
number of factors that will be common to
neighborhoods.
‣People’s interests, lifestyle and psychographic
characteristics and social image can be predicted by
looking at where they live.
‣Direct marketers should treat these characteristics as
a “statistical probability”.
Using External Databases
19
Creating a Geo-demographic Database

‣2 major analytical processes:

‣Data Reduction
‣ We don’t want the variables to be associated.
‣ Factor Analysis
‣Data is Combined into Segments
‣ We try to maximize the differences between groups, while minimizing the
differences within groups.
‣ Cluster Analysis.
Using External Databases
20
Lifestyle Databases
‣Data is collected at an individual level via commercially
operated surveys and product registration forms.
‣Four broad categories of data collected:
‣Name and address information
‣Data relating to the purchase of products and services
(known as ‘lead-to-purchase’ data)
‣Demographic and socio-economic information
‣Values and lifestyle (VALS) information.
‣Companies can also sponsor bespoke questions:
‣‘Would you be interested in supporting efforts to save
endangered species?’
‣The sponsor then has access to positive responders, usually
with a period of in-built exclusivity.
Using External Databases
21
VALS Data

‣Product needs are often “related to” or even “caused

by” lifestyle or demographic factors.
‣Internal data is often more useful in segmentation
processes
‣In situation where internal data needs an
enhancement external lists can be used for
segmentation purposes.
Using External Databases
22
Data Fusion

‣Data fusion is the combination of information on

individuals or groups using data from different
sources to present a more complete picture.
‣There are two key areas of data fusion:
‣Combining geo-demographic and lifestyle data
‣Combining market research databases with lifestyle data.
Using External Databases
23
Virtual Customers

Geodemographic-clusters- Lifestyle-clusters-

Match- ‣Building up “virtual

customers”.
Geo6lifestyle-clusters- ‣The data sources are:
‣Electoral Register
‣Lifestyle surveys
Assign-client’s- ‣Product registrations
Match- data-to-each-
cluster- ‣Census data
‣Shared databases.

Figure'4.4!!Building-up-virtual-consumers-
Using External Databases
24
Combining Market Research with Lifestyle Data

‣Research data and databases are either joined or

fused.
‣Joined:
‣Literally looking for duplicates of the same person in each data
set.
‣Fused (T-cube):
‣Linking together data profiles to create virtual profiles.
‣There is a great potential. But the management
process and the analysis of mass data set created
might produce some difficulties.
Using External Databases
25
T-Cube Data
Client'marke,ng'database,'or'
geodeographic/lfestye'data'

Market'research'data,'
e.g.'Target'group'index'

Figure'4.5!!T cube&data&
 26

USING EXTERNAL DATABASES

Issues and
Trends in
External
Databases
PART 3
Using External Databases
27
Strengths and Weaknesses of Geo-demographic Data

GEODEMOGRAPHİC,
Strengths, Weaknesses,

DATABASES,
Census&is&100%&sample& Only at OA/postcode level
No&bias& Only geodemographic data
‘Official’ survey more accurate? Modelled income only
Whole basis is geographical, and therefore ideal for Census records decay by 12
direct targeting years at end of life

Strengths( Weaknesses(
LIFESTYLE(DATABASES(

Individual data Partial coverage only

Behavioural/attitudinal data as well as Records biased to people who
geodemographic data like incentivised surveys
Claimed behaviour anecdotes
Income is actual not modelled
suggest quality is variable
Claimed update at tens of thousands of records per
Not based on geography
day
Consumers worried about data
protection

Table&4.3!!Strengths)and)weaknesses)of)geodemographic)versus)lifestyle)products))
Using External Databases
28
Issues in Company Usage of External Data

‣The quality of external data use is highly related to

computer technologies.
‣In companies that use external data there is a great
need for outsourcing and analysis support
Using External Databases
29
Recent Trends in E-Mail List Usage

Due to their low costs more companies are looking

for e-mail lists.

2 3 As e-mail lists are not detailed as external

geo-demographic and lifestyle databases, companies
fall behind in segmentation and targeting applications.

4 E-mail lists may be out of date.

1 E-mail lists may include “give-out-to-anyone” mail

addresses.