Content uploaded by Cristina Ziliani
Author content
All content in this area was uploaded by Cristina Ziliani on Mar 22, 2016
Content may be subject to copyright.
from 15 per cent in 1978. In Italy,
promotional spending increased by 10
per cent in 2003 and even further in
2004 and 2005, following a dramatic
slump in consumption (–2.4 per cent)
and retailers’ efforts to sustain sales.
The increased economic relevance of
promotions is not necessarily
accompanied by results: according to
Information Resources, in Italy, despite a
⫹1.6 per cent increase in promotional
RELEVANCE OF PROMOTIONS
Promotions are used extensively in
marketing, and promotional expenditure
has been slowly but steadily increasing
worldwide over the past 20 years.
1
Highly competitive market places compel
marketers to increase promotional
spending: in the USA, fast moving
consumer goods (FMCG) manufacturers
invest 25 per cent of their marketing
budget in consumer promotions,
2
up
䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00 Vol. 14, 3, 249–259 Journal of Targeting, Measurement and Analysis for Marketing 249
Target promotions: How to
measure and improve promotional
effectiveness through individual
customer information
Received (in revised form): 20th October, 2005
Cristina Ziliani
is a senior research associate in retailing and a lecturer in direct, database and internet marketing at the Universita`degli
StudidiParma,Italy.
Abstract This paper proposes a new approach to measuring the effectiveness of
consumer promotions, enabled by individual customer information collected via loyalty
cards. The paper begins by discussing the complexity associated with measuring
promotional effectiveness and the limits of current methodologies, then sets out to
demonstrate that a customer-focused approach to promotion measurement — as
opposed to the current product-focused perspective — can expand the measurability of
promotional effects in three dimensions:
• objectives — in that not only sales volume and market share effects can be tracked,
but also customer-specific and assortment-wide effects;
• time — in that promotional effects can be monitored for longer periods of time and
the fulfilment of long-term goals can be measured;
• space — thanks to geo-marketing analysis, sense can be made of the variability of
promotional response in different retail markets, channels and outlets.
This allows for better promotion planning and marketing effectiveness, with
consequences that extend beyond economic benefits for the retail company itself and
reverberate on relationships with suppliers and cooperative promotion decisions.
Cristina Ziliani
Universita`degliStudidi
Parma
Facolta`diEconomia
Via J. F. Kennedy 6, 43100
Parma, Italy
Tel : ⫹39 (0)521 902012;
Fax: +39 (0)521 032393
e-mail:
cristina.ziliani@unipr.it
measurement, demonstrating that it can
overcome the limits of current
‘sales-based’methods, and expand the
number of measurable objectives.
Finally, the paper will suggest
implications that the adoption of a
customer-focused approach to
promotions might have for retailers,
suppliers and vertical relationships; these
implications provide stimuli for further
research.
To address these goals, the following
research methods were combined:
•theoretical discussion of the complexity
intrinsic to promotion measurement,
based on an extensive literature review;
•interviews (from June to October
2004) with eight major Italian FMCG
retailers and manufacturers, and with
four market information multinationals
—ACNielsen, Information Resources,
Catalina Marketing Inc. and
DunnHumby —to identify limitations
of current methodologies employed for
the measurement of promotional
effectiveness;
•development and testing of new
customer-focused measures on a real
marketing database, provided by a
major Italian retailer, and design of the
blueprint for a decision support system
aimed at assisting retailers in their
choice of category and brand for
promotional purposes;
•discussion of preliminary results with a
selected group of five Italian retail
managers.
pressure, and ⫹0.8 per cent increase in
markdown, in the supermarket industry,
the result in terms of sales has been a
disappointing ⫹0.3 per cent (Table 1).
Despite their diminishing returns,
promotions cannot be disposed of, as
they are permanent competitive
‘weapons’ for retailers. Indeed, retailers
and manufacturers are likely to increase
promotional spending in the future.
Moreover, promotional strategies are
evolving, due to the diffusion of a
loyalty/retention culture among retailers.
The new focus on retention shifts
promotions from outside the point of
sale to the inside, from mass media to
direct media, from one-size-fits-all to
cluster marketing, from traffic building to
share-of-wallet building efforts.
3,4
Promotional plans grow in complexity, as
massive adoption of loyalty schemes
makes it necessary to integrate continuity
promotions with short-term price
promotion.
Discussion on how to make this
investment more effective is therefore
necessary, and this paper aims to
stimulate it.
The first goal of this exploratory
research is to discuss the complexity of
measuring promotional effectiveness and
analyse the ‘defects’ of methodologies
currently in use in the industry, such as
those developed by ACNielsen and
Information Resources Inc.
The second and main objective of the
research is to propose a new
customer-focused approach to promotion
250 Journal of Targeting, Measurement and Analysis for Marketing Vol. 14, 3, 249–259 䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00
Ziliani
Table 1: More pressure does not mean more sales (Italy 2004)
Change over Change over
Hypermarket previous year (%) Supermarket previous year (%)
Promotional pressure (%) 30.8 ⫹1.5 21.1 ⫹1.6
Markdown (%) 24.3 ⫹0.5 24.5 ⫹0.8
Effectiveness 59 ⫺0.1 53 ⫺0.3
Source: Information Resources
advertising and trade promotions, are
taking place; increasingly, several types
of promotions (continuity, price and
targeted ones) are run at the same
time, and respective effects are difficult
to isolate;
•effectiveness is a ‘jigsaw’,inthatit
results from the combination of the
effects of diverse promotional elements:
promoted category/brand, target,
promotional tool, type of incentive,
depth of discount, redemption effort,
timing,media,message,testingandso
on.
Given such complexity, the problem of
measuring promotional effects has usually
been solved by limiting the analysis to
the most immediate and measurable
aspect of the phenomenon: the so called
‘sales effect’—that is the increase in
sales volumes (and values) generated by
the promotion.
Three common methodologies are
used —alone or combined —to
measure the ‘sales lift’:
•incremental sales;
LIMITS OF CURRENT
MEASURES OF PROMOTIONAL
EFFECTIVENESS
According to Busacca and Mauri,
5
promotional effectiveness is defined as
‘thedegreeoffulfilment of the goals
pursued by the company by means of a
specific promotional action over a given
period of time’.
Measuring the degree of fulfilment is
difficult for the following reasons:
•there can be more than one goal to a
promotion (Table 2), and goals may go
beyond a simple sales lift on the
promoted item;
•such goals are seldom clearly stated,
due to the tactical nature of
promotions;
•promotional effects before (lead)and
after (lag/stock up) the actual
promotional period can pass unnoticed;
•promotions display intrabrand,
interbrand, intracategory, intercategory
and store switching effects that escape
conventional measures;
•consumer promotions are run while
other marketing activities, such as
䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00 Vol. 14, 3, 249–259 Journal of Targeting, Measurement and Analysis for Marketing 251
Promotional effectiveness through individual customer information
Table 2: An overview of promotional objectives
Short term Long term
Sales-related objectives Increase sales and turnover Expand market share
Sustain volumes Consolidate market share
Regain lost market share Expand brand's customer base
React to competitor's move Increase exclusive use of brand
Get rid of excess inventory Create entry/mobility barriers
Customer-related objectives Stimulate first trial of product Build loyalty
Stimulate repurchase Permanently lift consumed quantity
per customer
Increase quantity usually purchased Increase exclusivity of use
Increase frequency of purchase Suggest new uses for product
Focus customers' attention on brand Modify brand positioning
Up-sell to larger size of product Support brand image
Increase stock in home Expand share of wallet
Stimulate impulse buying –
Collect information about consumers –
Trade-related objectives Increase trade inventory of brand Increase number of stores carrying
brand
Push product listing by retailers Increase category share
Increase visibility in store –
Source: Adapted from Busacca and Mauri (1994)
the short term, typically over three- or
six-week periods: Such time spans are
standardised and are often shorter than
the real repurchase cycle for the item,
therefore distorting estimates of lead
and stock-up effects and failing to
capture medium- and long-term effects
of promotions. This is a ‘short-sighted
mistake’made by data users
(manufacturers and retailers looking at
the data), rather than a pitfall of the
methodologies. Market information
companies, if requested, can supply
customers with longer time series for
analysis;
•Intrabrand, interbrand, category and store
switching effects almost entirely escape
measurement: This is the most notable
defect of sales-based methods, and the
most relevant for retailers, whose goal
with promotions is to sell an
assortment, rather than single stock
keeping units, brands or categories.
Switching effects can only be tracked
and understood when there is a
customer perspective to promotional
analysis, by looking at shopping baskets
over time.
The ‘variationofmarketshare’method,
building on the incremental sales
method, displays the same pitfalls and
will not therefore be discussed here.
The ‘redemption’method, typically
used for coupons, is similarly faulty, in
that the mere amount of redeemed
coupons is no evidence of effectiveness
per se. For example, a redemption rate of
40 per cent would, at first glance, lead to
the assertion that the promotion was a
success. But what happens when a closer
look at the purchase records of those
customers who took advantage of the
promotion reveals that it was mainly
customers loyal to the promoted brand
(those who would have purchased
anyway) who redeemed the coupon,
while no new customer was attracted to
•variationofmarketshare;
•redemption.
The incremental sales method is the
commonest: ACNielsen and Information
Resources established long ago their own
methodologies to calculate the ‘baseline’
(ie sales that would occur anyway, not
attributable to specificmarketing
activities) and hence determine
promotional lift. Following interviews
with Italian companies employing those
methodologies, and with ACNielsen and
Information Resources, who developed
them, such methods were found to suffer
from several defects, as listed below.
•Assumptions regarding the actual presence of
apromotion: Based on the detection of
price variations over a four-week
period, analysts assume that a
promotion is being run when a 10 per
cent difference (for ACNielsen, 5 per
cent for Information Resources) in
price level per item is spotted. The
two thresholds suggest that the two
companies account for a different
number of promotions, resulting in
different ‘universes’.Moreover,
promotions are not monitored in all
stores belonging to ACNielsen and
Information Resources panels, hence
limiting the extensibility of the results
to the market (market share method).
•Absence of ‘promotion-free’weeks for certain
categories: The baseline calculation
requires sales data from
‘promotion-free’weeks. However,
some categories where promotional
pressure is very high never experience
a promotion-free week during the
year, forcing analysts (when calculating
the baseline at category level) to make
up a ‘mock’promotion-free week by
adding sales data from different
promotion-free weeks for each item
belonging to the category.
•Post-promotion effects are only monitored in
252 Journal of Targeting, Measurement and Analysis for Marketing Vol. 14, 3, 249–259 䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00
Ziliani
discussion with a group of five retail
marketing managers.
Overcoming the limits of
‘sales-based methods’
We mentioned earlier that in addition to
sales-related objectives, a wide range of
customer-related and trade-related
objectives can also be envisaged for
promotions. As far as customer-specific
objectives are concerned (Table 2),
loyalty data allow for a substantial
advancement in measurability. Not only
isthesamplesignificant (the database
often contains more than one year of
purchase behaviour data for thousands of
customers), but data reflect actual
behaviour and are timely —a relevant
fact, especially for promotion planning, if
one considers that in FMCG preferences
change quickly.
Most notable is the fact that, by
identifying customers who took
advantage of a promotion, it is possible
to quantify fulfilment of specific
objectives, such as the extent to which
the promotion resulted in first trial,
repurchase, sales subtraction from other
brands/categories/stores and so on, as
shown in Figure 1. Here, the
information provided by the incremental
sales method (the ‘lift’)iscomparedwith
that provided by the decomposition of
the lift according to the customer
segments who took advantage of the
promotion (see pie chart in Figure 1).
The generic increase in sales can be
attributed to new customers (1 per cent),
switchers (48 per cent), customers loyal
to promoted brand (35 per cent), light
consumers who were persuaded to buy
more (10 per cent) etc, providing the
retailer and the manufacturer with a
more satisfactory understanding of the
promotion’s effects. It is worth noting
that retailers can keep such additional
insight to themselves, or share it with
the brand and no light consumers were
persuaded to speed up their repurchase
cycle and buy more? Is the promotion
still regarded as effective? The
importance of setting explicit
promotional goals beforehand, and
expressing them in terms of customer
behaviour, not simply in terms of
sales/volumes/number of coupons
becomes evident.
A NEW CUSTOMER-FOCUSED
APPROACH TO PROMOTION
MEASUREMENT
When they were first introduced,
scanners produced a major breakthrough
in marketing and sales measurement.
Before scanners, less reliable, patchy ex
fabrica data were used. Almost 30 years
later, it is not surprising that the
dominating approach to promotional
measures is sales driven; nonetheless, a
new era for marketing measurement is
coming, ushered in by a different
technology: the customer database.
The third goal of this paper is to
illustrate how individual demographic
and behavioural customer information,
such as that collected by loyalty cards,
can be used in conjunction with scanner
(sell-out) data to:
•overcome the limits of current
sales-based methods;
•measure the degree of fulfilment of
customer-specific, long-term and
assortment-wide promotional
objectives;
•create a knowledge repository which, if
leveraged,canhelptodesignbetter
promotions.
The following discussion is based on
analysis of the customer database of an
Italian retail chain, and on the
outcome of interviews. The findings
were validated by means of a
䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00 Vol. 14, 3, 249–259 Journal of Targeting, Measurement and Analysis for Marketing 253
Promotional effectiveness through individual customer information
manufacturers’assumptions over purchase
frequency for their products —typically
derived from market research —differ
from actual behaviour. Moreover,
frequency is a ‘learned behaviour’and it
can differ widely from store to store
depending on the promotional activity
adopted (ie the promotional pattern in
certain stores might have trained
customers to ‘wait for sales’).
Capturing switching effects
As mentioned above, retailers are most
interested in understanding the
cross-elasticity of product demands, as
they are continuously using promotions
to support sales of the whole assortment
of goods. By looking at shopping
baskets, it is possible to monitor the
impact of promotional actions not only
on the promoted items, but most notably
on other brands, categories, customer
segments and stores. It is possible to
establish how closely one
product/brand/category/store is a
substitute for another, with clear
implications for pricing, merchandising,
store layout and future promotions.
Table 3 shows substitution dynamics in
the partner manufacturer, depending on
the promotional agreement and the
relationship.
Sales-based methods typically measure
short-term promotional effects. With the
new approach, the group of customers
who bought during the promotion can
be monitored over a much longer period
of time. This enables companies to verify
the long-term effects of the promotion:
did consumers only buy the promoted
good once, or did they repeat the
purchase after the promotion was over?
Have they become loyal to the
promoted item or did they go back to
their favourite brand, and so on (see
Figure 2). Here, the lines show how the
promoted items performed over a
three-month period after the promotion,
compared with the category they belong
to: the increase in category share ‘held’
in the long term, in terms of volume,
value and number of customers.
Individual customer purchase histories,
moreover, enable the calculation of ‘real’
product purchase cycles, which can be
used to determine the appropriate time
span for measurements of effectiveness,
and better estimates of lead and lag
effects. In fact, it is often found that
254 Journal of Targeting, Measurement and Analysis for Marketing Vol. 14, 3, 249–259 䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00
Ziliani
Figure 1 Measuring customer-specific objectives ‘hidden’under the veil of promotional lift
0
50
100
150
200
250
300
Regular sales Promotional sales
48%
10%
6%
35%
1%
Switchers
Light consumers
Loyals to other brand
Loyals to brand A
New customers
segments. For example, a retailer might
want to explore how the ‘top’segment
of the customer base reacts to
promotions, or how the ‘deal conscious’
segment —that is, the group of
customers who are most interested in
promotions —welcomes a specific
action.
Geo-marketing analysis can help to
track the behaviour of customers who
switch stores within the same chain for
promotional reasons, and, most notably,
itcanalsohelptoputtheword
the mayonnaise category over a one-year
period. In the right-hand column it is
possible to read the average category
share of each brand during the year. The
other columns show what happens to
each brand’s share when one or more
brands are promoted —that is, which
brands are more affected when others
become more attractive to customers.
This intracategory switching analysis,
based on shopping baskets, can be
extended to include substitute categories
and cross-tabulated with customer
䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00 Vol. 14, 3, 249–259 Journal of Targeting, Measurement and Analysis for Marketing 255
Promotional effectiveness through individual customer information
Figure 2 Tracking longer term effects of promotions
Table 3: Measuring variation of intracategory brand share over one year
Category share
Category share Category share when Category share
when SMA when KRAFT SMA + KRAFT over year
Brand switching is promoted is promoted are promoted (mean)
SMA share 50.5 35.0 49.8 42.0
KRAFT share 20.0 44.3 26.0 31.0
CALVE share 29.5 20.7 23.7 27.0
Source: Catalina Marketing Inc.
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
May June July August September
Share (value) Share (volume) Share (customers)
Share of promoted items over time
me-too decisions (let’s promote the same
category as the competitors...) and at the
same time serving the retailer’sown
objectives, not only the partner
manufacturer’s.
When a customer-focused approach to
promotions is adopted, new measures
and indicators that help to refine
decisions can be constructed. As an
example, a prototype of category/brand
choice decision support system for
promotions was created (Figure 3); this
reveals differences in potential
effectiveness for categories that at first
sight seemed equivalent. The value of
information such as the percentage of
‘gold’customers who buy the category,
the intensity of brand switching, the
promotional proneness of customers
shopping within the category is
self-evident. This information is all made
available by the new approach.
Effectiveness can be increased ex ante
also via segmentation and appropriate
targeting of the customer base:
segmentation criteria should be defined
according to promotional objectives. If
the goal is to increase share of wallet
or brand loyalty, for example, customers
should be segmented and targeted
accordingly. One of the most
interesting applications, in this respect,
is segmentation of customers according
to promotional behaviour, an exercise
that Tesco has been carrying out
successfully on their customer database,
as DunnHumby explained during the
interview. Needless to say, to exploit
this segmentation to the full,
communicate directly with selected
segments and target them with ad hoc
promotions, there needs to be a system
in place. Loyalty card databases can
help, if information such as address,
telephone number and e-mail address
of customers have been captured and
updated over time. For a discussion of
European retailers’adoption of targeted
‘effective’into perspective. Indeed, the
‘degree of fulfilment’of any promotional
objective can be labelled as satisfactory
(or not) only if it is compared with the
promotion potential of the category and
brand —a concept well known to
retailers —but also the promotional
potential of the area (the number,
spending levels and preferences of
customers in the target area affect
promotional results). It emerged from the
interviews that so far this geographical
aspect of promotional potential has not
been taken into account for promotional
decisions and evaluations.
Creating more effective promotions
by leveraging the promotional
knowledge base
This returns to the statement that
effectiveness results from the combination
of diverse promotional elements:
promoted category/brand, target,
promotional tool, type of incentive,
depth of discount, redemption effort,
timing,media,message,testingandso
on. In order to improve the design of
future promotions, it would therefore be
very useful to keep track of past
promotional decisions regarding each
element, and their impact/result, by
appending information to the database.
This study found that, due to the tactical
nature of promotions, their frequency
and the sheer quantity of data that needs
to be processed and set aside,
promotional information other than sales
volumes is seldom fed into marketing
information systems.
Effectiveness, however, can be ‘built’
into promotions by learning from
accumulated knowledge. Take, for
example, the decision regarding what
category to run the promotion in: the
customer database can serve as an
excellent information base to choose the
appropriate category, avoiding the trap of
256 Journal of Targeting, Measurement and Analysis for Marketing Vol. 14, 3, 249–259 䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00
Ziliani
hand, the new customer-focused
approach introduces retailers to a
typically industrial way of reasoning —
customer-focused objectives —while the
appreciation of intrabrand, interbrand,
intracategory and intercategory effects
broadens the manufacturer’sviewofthe
market, by definition narrower than that
of the retailer. From this viewpoint,
then, the two marketing visions might
move closer. On the other hand,
retailers’objectives are different from
those of manufacturers’—consider the
motivation to sell a range, rather than
specific brands, to retain best customers,
or to expand share of wallet; once
retailers are better enabled to measure
such objectives, they will select
promotions accordingly, refusing to run
and cooperate on promotional activities
that hamper fulfilment of their own
goals. A clearer vision of promotional
results can shake long-established
partnerships and conflict might become
promotions and direct media see Ziliani
and Bellini (2004).
6
Impact on vertical relationships
With this information, and new measures
and tools, retailers can take a more
mature and discerning approach to
promotional decisions, especially when it
comes to choosing between deals offered
by different manufacturers. Data about
the category/brand potential and past
promotional results can be quickly
accessed (as shown in Figure 3) and
negotiation influenced accordingly. It is
likely that, in the future, a retailer’s
evaluation of prospective partners will
incorporate elements such as the ability
to discuss and set customer-specificand
long-term objectives.
It is uncertain, however, whether the
availability of customer information and
measurability of customer-specificand
long-term objectives will bring retailers
and manufacturers any closer; on the one
䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00 Vol. 14, 3, 249–259 Journal of Targeting, Measurement and Analysis for Marketing 257
Promotional effectiveness through individual customer information
Figure 3 A customer focused decision support system to assist choice of category and brand for
promotions
Isit st orable ?
Whatisthe promotional pressureon the categoryin the market?
Whatisthecategoryrolefortheretailer?
Howmanycustomers purchasethe category?
Whatar e these c ustomers worth for t he store?
Purchasefrequency
Purchase value
Numberofitems purchased
Average cumulat ed value of purch ases in the categ ory per custo mer over 1 yera
Share of categor y leader
,bran ds shares ( value a nd volume)
Numberofloyal customersper brand
Pronenesstobrand switching
Penetration of c ategory among gol d customer s
Retailers own cu stomer segm ents: such as go ld, silver , tin
Penetrationofeach brandin different customersegments( gold,silver, tin)
Demographicsofeach brandscustomer base
Mayonnaise
Promotionalresponsivenessofeach brandscustomer base
Is it storab le?
What is the promotional pressure on the category in the market?
Whatisthecategoryrolefortheretailer?
How many customers purchase the category?
What are t hese customers worth for the store?
Purchase frequency
Purchase value
Number of it ems purchased
Average cumulated value of purchases in the category per customer over one year
Share of categor y leader
brands shar es ( value and volume)
Number of loyal customersper brand
Proneness to brand switching
Penetration of category among gold customers
Retailers ow n customer segments: such as gold, silver, tin
Penetration of each brand in different customer segments (gold, silver, tin)
Demographics of each brands customer base
Mayonnaise
Promotional responsiveness of each brands customer base
•Promotions drain increasing amounts of
financial resources and become more
complex, while their returns are
diminishing; retailers and manufacturers
will therefore strive for ways to
improve effectiveness and to obtain
additional resources by making loyalty
investments pay for themselves.
•Atthesametime,theretentionculture
and the availability of individual
customer information bring
customer-specific promotional
objectives to marketers’attention,
urging them to rethink generic sales lift
goals.
•The discovery of customer
heterogeneity and the ability to
carefully segment and target the
customer base, coupled with efficient
direct media such as e-mail, SMS and
electronic coupons, create conditions
to expand the variety of promotional
tools and actions.
•In this scenario, a customer-focused
approach —as opposed to the current
product-focused perspective —
increases the possibility of measuring
promotional effectiveness in three
dimensions:
—objectives, in that not only sales
volume and market share effects can
be tracked, but also increasingly
important customer-specific ones
(supplier viewpoint) and
assortment-wide ones (retail
viewpoint);
—time, in that promotional effects can
be traced for longer periods of time
and the fulfilment of long-term
goals can be measured;
—space: thanks to geo-marketing
analysis, sense can be made of the
variability of promotional response
in different retail markets, channels
and outlets.
•One could argue that consumer panel
research such as Nielsen Home scan
canservethesamepurposebut,
harsher. Research is therefore needed to
further explore the impact of improved
promotional measurement on vertical
relationships.
CONCLUSIONS AND STIMULI
FOR FURTHER RESEARCH
Research on retailers’use of loyalty card
information across Europe began several
years ago. Today, evidence shows that,
on the one hand, retailers declare that
customer information is vital for
competitive advantage; on the other
hand, they hesitate to put data to any
use.
7,8
Retailers are seeking ways to make
loyalty investments pay for themselves: in
the literature, at least, there seem to be
many attractive options. With the often
cited exception of Tesco in the UK,
however, applications of individual
customer information in retailing are
rather conservative, limited to improving
and running the loyalty scheme, direct
communication and location analysis, as
found by the survey of the Italian market
in 2003 and confirmed on a European
scale.
9
Several avenues are still to be
explored, where scope for value creation
exists, such as segmentation based on
shopping baskets, ranging stores
differently, new product/service
development, cross-selling across
channels.
In order to move from mass to
information-intensive marketing,
however, retailers should not dissipate
their efforts in many directions, but focus
on a few priority areas of intervention,
depending on their marketing mix and
competitive position. This paper suggests
that one such area is the measurement of
promotional effectiveness, and the
consequent improvement of promotional
decisions.
The following points emerging from
the research would benefitfromfurther
exploration:
258 Journal of Targeting, Measurement and Analysis for Marketing Vol. 14, 3, 249–259 䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00
Ziliani
•The expansion of promotional
experimentation and variety, however,
is likely to be limited by cost factors,
complexity of promotional plan
management, shortage of marketing
analysis skills, cultural factors such as
the long-established practice of
entitling buyers, not the marketing
function (where knowledge about
customers resides), to decide which
manufacturer’s promotional proposal to
accept.
References
1 Bell, D. R., Chiang, J. and Padmanabhan, V. (1999)
‘The decomposition of promotional response: An
empirical generalisation’,Marketing Science,Vol.18,
No.4, pp. 504–526.
2 Raghubir, P., Inman, J. J. and Grande, H. (2004)
‘The three faces of consumer promotions’,California
Management Review,Vol.46,No.4,pp.23–42.
3 Hawkins, G. (1999) ‘Building the customer specific
retail enterprise’, Breezy Heights Publishing, New
York, NY.
4 Ziliani, C. and Bellini, S. (2005) ‘Le cible´marketing
en Italie: Tendances et pre´visions de de´veloppement’,
Revue Française de Marketing, Vol. 203, Juillet, pp.
3–5.
5 Busacca, B. and Mauri, C. (1994) ‘L’efficacia
dell’azione promozionale al consumo’,Finanza,
Marketing e Produzione, No. 3, pp. 45–84.
6 Ziliani and Bellini (2004) op. cit.
7 Cuthbertson, R. and Laine, A. (2004) ‘Theroleof
CRM within retail loyalty marketing’,Journal of
Targeting, Measurement and Analysis for Marketing,Vol.
12, No. 3, pp. 290–304.
8 Stone, M., Bearman, D., Butscher, S. A., et al.
(2004) ‘The effect of retail customer loyalty schemes
—Detailed measurement or transforming
marketing?’,Journal of Targeting, Measurement and
Analysis for Marketing,Vol.12,No.3,pp.305–18.
9 Cuthbertson and Laine (2004) op. cit.
generally speaking, such panels are
designed to answer
manufacturer-specific questions
regarding promotions, are based on
statistical generalisation and, in markets
such as Italy, where the retail market is
highly fragmented, only a few retailers
can find answers regarding their own
customer base in the panel. Most
banners go unrepresented. Finally,
panel information cannot be used
directly for targeting one’sown
customer base.
•Such expanded possibilities will have
an impact on vertical relationships:
improvements in the three areas above
reverberate on the retailer’s
promotional decision-making process,
affecting choice of category, brand,
timing and partner manufacturer for
promotions. As a result, the
conflict/cooperation balance in the
retailer’s relationship with suppliers can
change. It is likely that, in the future, a
retailer’sevaluationofprospective
partners will incorporate elements such
as ability to discuss and set
customer-specific and long-term
objectives. Manufacturers who are
quick to adopt the new language of
promotions and are willing to
incorporate customer-specificand
assortment-wide objectives and ex-post
measures in their promotional
agreements (and also category
management agreements) with retailers
mightgainanadvantage.
䉷Palgrave Macmillan Ltd 1479-1862/06 $30.00 Vol. 14, 3, 249–259 Journal of Targeting, Measurement and Analysis for Marketing 259
Promotional effectiveness through individual customer information
