Module II

The Database
 Marketing database
• Form of customer records
• Records sales, communication history, personal
details
• a database is a collection of data records in a list
that can be manipulated by software
• a data record is the entire set of information that
is associated with one customer
• a data field is one item of data within a record:
for example, a name, address, the number of a
type of product bought
 A marketing database
• list of customers’ and prospects’ records that
enables strategic analysis, and individual
selections for communication and customer
service support. The data is organised around
the customers.
 Types of data in a marketing database
• Primary data, such as names and addresses of customers,
details of products /services’ your organization offers, pricing
details, campaign details and definitions of various channels
of distribution
• Secondary data, which is data used to qualify primary data,
such as demographics, lifestyle information, geographical
profiles or levels of penetration into markets
• Performance data to record how your customers have
responded, what they bought, how much they spent and to
which campaigns they responded
• External data, covering everything rented or bought in lists to
data from various agencies that can enhance your base data
 Principles of data collection
• Data should be split into ‘essential now’ and
‘possible future use’.
• Data should allow ease of sourcing and
updating.
• The cost of raw data, for example external
bought in data, must be offset against benefits
• ‘only hold data that is required for your
strategy. If you don’t use it, don’t collect it and
if you collect it, use it.’
Typical customer record
 Data sources
• Internal sources-Direct customer contact/indirect customer contact
Direct customer contact- retailers, etc
- Records held by anyone within the company who deals with
customers
• Retail outlets, agents, dealers
• Guarantee forms
• Invoices
• Enquirers, past, present and future
• People who visit your stand when you are at any exhibitions or
shows
• Competition entrants
• Responders to sales promotions
• External sources
 Data sources
• Companies with no direct contact- FMCG etc
- List purchase
- Data building scheme- through sales promotion
- New contact channels-
• product registration documents, product warranties
• credit card details
• subscription details
• questionnaire responses
• in-store offer details
• requests for product information
• events/ promotions requiring response
• established direct channels
 Database management

1 2 3 4
Manage the Manage the Manage the Manage the
data sources data entry database applications
Step 1- Managing the data sources
• Many no. of customer related databases from different functional
areas such as sales, marketing, accounting
• Different databases might record different customer related data
such as opportunities, campaigns, enquiries, deliveries, billing etc.
• Can be about individual customers, customer cohorts, segments
• The Key for direct marketing is to organize data around ‘Customer’
and not ‘other factors’
• Co-operations across departments for organizing data around
customers
• Regular audit of data quality
• Training for front end staff in capturing data
• Regular updating of records
 Step 2- Managing the data entry
• Verifying that the data has been put in
Verification • approach is to select records at random for visual
inspection and correction
• is checking the accuracy of personal and product data provided by
the customer
• checking product source codes for invalid codes against an internal
master list
Validation • Audits
• Accuracy checks
• Sanity checks

• is the act of ensuring that the database does not contain duplicate records ofthe
same customer.
Deduplication • Customer data should not be duplicated like Micheal Sethi is written as Micheal S
and Micheal Sethi though both are same customers

Merge purge • merging two files – perhaps two external, or one external and your own internal
database – into one
 Step 3: Manage the database
• In house or outsourced
• Managing interface between Marketing
department and IT department – Managing
conflicts between IT and Marketing objectives,
technical specifications, hardware, software
• Data management over time – Updating ,
Auditing, Archiving
 Step 4: Manage the applications
• Selection : Selection of customer list based on
marketing objective
• Outputs : Outputs are those fields which are
transferred from the database to the
communication material going to the customer
like salutation, address etc.
• Managing campaign responses : Communication
campaign code (Unique Reference Number, URN),
Capturing customer response (Yes/No) to the
campaign objective
 Problem
• You are a large car dealer selling new as well
as old cars. Recently you have noticed that
customer traffic is dwindling. Social media
opinion shows that customers are losing trust
and are complaining about poor service.
He decides to implement direct marketing to
salvage the situation. Suggest as to how he
can go about the same.
Customer database and segmentation
• Customer database- volume data and
customer information- helps in segmentation
• Use purchase data – to analyze consumer
needs –according to consumer value and
product needs
• Use to create “customer pictures”
 Segmentation applications
• Use purchase behavior to segment by
customer value (also, use current purchase
behavior to forecast lifetime values).
• Use purchase behavior to segment by product
need, and to assess product/market strategy.
• Use profile data to relate to campaign
response (ROI driven marketing).
• Use profile data to target new customers
accurately.
 Using purchase data to segment by value

• Target only few valuable consumers

• Uses Paretos principle- 80-20 rule -Twenty per cent of your
customers will provide you with 80 per cent of your profits.’
• Identification of best customers
• Companies can analyse their customers’ purchase data in
order to calculate how much each customer is likely to be
ultimately worth to the company. This forecast is known as
the ‘customer’s lifetime value’.
• By predicting how much gross profit we will make from
customers, we can then construct a budget, on a per
customer basis, which clearly tells us how much we can afford
to spend. This calculation is called the ‘allowable marketing
spend per customer’, and is extremely important in direct
marketing practice.
 Segmenting according to customer need

• different benefits obtained by different sets of

consumers from the same product.
• Also uses behavioural and psychographic data
to understand consumer needs
Segmentation at Harley Davidson Bike
lifestyle needs
• Tour Gliders – ‘My bike is made for comfort rather than
speed’
• Dream riders – ‘Most of the time my bike is parked. I
mainly use it for short trips
• Hard core – ‘Some people would call me and my friends
outlaws. I think its true that real men wear black’
• Hog Heaven – ‘When I ride, I feel like an old wild west
cowboy’
• Zen riders – ‘I like the attention I get when I am on my
bike’
• Live to ride – ‘I love to ride long distances: to me 500
miles is a short trip’
 Maximising response on a per campaign basis

• use modelling techniques to identify their most

‘responsive’ customers.
• This allows us to allocate the marketing budget on a
per campaign basis such that return on investment
(ROI) is maximised.
• Helps to identify customers who are most likely to
respond in view of limited budget
• use predictive statistics, to forecast which individual
customers are most likely to respond
 Targeting new prospects based on existing customer
descriptions: profiling

• Provided the new markets are looking for the same

benefits from the product as existing markets:
• your new customers will have a similar profile to your
existing customers
• Use modelling techniques
 Analytical techniques in Direct Marketing

• Two methods used – descriptive and predictive

• time series methods: make predictions about future
levels of customer response by projecting forwards
from the trend, cycle and seasonal patterns contained
in ‘past values’ of data. Quick and accurate
• Explanatory methods can predict customer response
(the dependent variable) by identifying the factors that
drive it (the independent variables), estimating their
effects and projecting these forward into the future.
more useful than time series analyses when looking at
the medium to long-term future
 Time series- Customer lifetime value (CLV)

• Explores the relationship between the gross margin

from customers and, the amount we can spend
marketing to them.
• Lifetime values are the sum of all future net incomes
expected from a customer, translated into present day
terms.
• CLV uses:
• the number of years a customer buys from a company
• the percentage of customers remaining loyal to the
company
• the amount spent per annum
Calculating allowable marketing spend
per customer
• The first is the amount we can spend to attract
a customer for the first time.
• This figure may well have to be greater than
the revenue from the first sale.
• The second is the amount we can spend on
existing customers in order to keep them with
us.
 Frequency, recency, amount and
category (FRAC)
• a way of assessing each customer on the database,
based on their purchase behaviour
• Frequency refers to the time elapsed between
purchases.
• Recency is the date of the most recent purchase.
• Amount is the average value of that customer’s
purchases.
• Category refers to that segment which the product
bought comes under.
• Helps to predict the likelihood of responding to your
next campaign
• Customers are scored according to FRAC data
• Then ranked
• Works on principles of
• Your next customer is likely to be the one who has just
bought from you
• A customer with many accounts is more likely to
purchase further products from you than one with few
or no accounts
• The more your customers have spent with you, the
more they are likely to spend with you in future
 Example – FRAC Scoring
Segment Decile (%) Score Marketing
(Based Decision
on Decile
Rank)
1 0-10 60+ Higher contact
(More
Responsive)
2 11-20 54-59
3 21-30 48-53
4 31-40 42-47
5 41-50 36-41
6 51-60 30-35
7 61-70 24-29
8 71-80 18-23
9 81-90 12-17
10 91-100 6-11 Lower contact
(less responsive)
 Explanatory analyses
• Cluster analysis
• Neural networks
• Regression
• CHAID-Chi-squared Automatic Interaction
Detector