Janus and the changing face of
pricing research
White paper
Authors:
1 / 21
Maureen Arink
Vicky Nef
Anne Favrelle
Presented at the ESOMAR Congress 2010
Nominated for best methodological paper
Rotterdam
Rotterdam
| Geneva
| Geneva
| London
| London
| New
| New
York
York
www.skimgroup.com
www.skimgroup.com
Table of contents
Introduction _______________________________________________________________________ 3
Belief 1: Price Elasticity to lowered prices is the same as price elasticity to raised prices. ___________ 5
Belief 2: The higher the standard of living in a country, the lower price sensitivity of consumers.______ 8
Belief 3: German consumers are very price sensitive; more price sensitive than consumers in other
European countries. _______________________________________________________________ 11
Belief 4: It is better to change your pack size than to change your shelf price if you want to increase
prices. __________________________________________________________________________ 13
Belief 5: Women are more price sensitive than men. ______________________________________ 14
Belief 6: The lower the personal involvement of a category, the higher the price elasticity. _________ 15
Belief 7: Price elasticity is higher with larger packs than with smaller packs _____________________ 17
Conclusion ______________________________________________________________________ 17
References ______________________________________________________________________ 17
The authors ______________________________________________________________________ 17
Appendix ________________________________________________________________________ 17
2 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Introduction
Pricing is one of the most powerful marketing levers. It represents the harvesting of all communication
and brand architecture investments; it also brings a golden opportunity to communicate what your brand
is worth vis-à-vis competitive offerings. Get your price right and you can maximize profit and strengthen
your brand - get it wrong and you can lose revenue, confuse consumers, or even start a price war.
It’s not surprising that pricing research never goes out of fashion. But ad-hoc research has its place great business decisions are made by combining relevant market data and an astute gut feeling for how
your market will react to pricing changes. Unfortunately gut feelings aren’t quick to develop, nor can they
be easily justified when challenged.
SKIM has carried out over 200 pricing studies over the last five years. Leveraging the pricing knowledge
generated across studies, we have created a database to show the consumer behavioral trends
between countries, categories and consumer types. Thus we have generated a base for testing gut
feelings, for building general wisdom when it comes to pricing decisions.
figure 1 The changing face of Janus
In this paper we share a few examples of the type of insight we can get from our pricing database. With
the 2010 ESOMAR Conference in Athens in mind, the stories we tell are introduced as seven common
beliefs about pricing. We take these seven common pricing beliefs and identify whether they should be
dismissed as myths, or whether there is any evidence of solid foundation behind them.
Belief 1: Shopper sensitivity to lowered prices is the same as shopper sensitivity to raised prices.
Belief 2: The higher the standard of living in a country, the lower price sensitivity of consumers.
Belief 3: German consumers are very price sensitive; more price sensitive than consumers in
other European countries.
Belief 4: It is better to change your pack size than to change your shelf price if you want to
increase prices.
Belief 5: Women are more price sensitive than men.
Belief 6: The lower the personal involvement of a category, the higher the price elasticity.
Belief 7: Price elasticity is higher with larger packages than with smaller packages.
3 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Before we start however, we would like to give a short explanation as to what we mean by price
sensitivity and price elasticity.
Price sensitivity describes consumer behavior. Consumers are sensitive to price changes if they switch
from one product to another as prices move – consumers are price insensitive if they remain loyal to one
product despite price changes.
When we think and talk about products, we rather use the term price elasticity. The elasticity of a
product is said to be -1.0 when a change in price of +1% leads in a change in volume sales of -1%. A
price elasticity of -2.0 infers that as price increases by 1%, for example, volume sales will decrease by
2%.
Throughout this paper we refer to price sensitivity when we are consumer, or shopper, focused and price
elasticity when we are product focused.
4 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Belief 1: Price Elasticity to lowered prices is the same as price elasticity to raised
prices.
Typically our clients come to us with some pricing related business questions. Often, we will design and
run a pricing study, deliver insights and a nice simulator to test any more scenarios they might think of
trying out. After sharing with interested parties and making some decisions, the research findings and
tools will be put on a handy server somewhere for reference. Hopefully, the brand team will also keep
top of mind the big picture that can be used in daily work: Product A has an elasticity of -1.5 and Product
B of -2.0, for example.
Stating that Product A’s price elasticity of -1.5 has an inherent assumption. It assumes that the elasticity
of A is symmetric – move price up 10% and you lose 15% volume; move price down 10% and you will
win 15%.
figure 1.1 The mythical symmetrical price elasticities
When asked, most marketing professionals will take this as an acceptable simplifying assumption. But it
is a big assumption. And we are now going to show, it’s a bad assumption.
Actually, when challenged to think more about it, most people would guess that when taking prices up,
you loose faster than you gain when taking prices down. Furthermore, academic literature on pricing
often refers to prospect theory. [ref 1]. ) Prospect theory is a cognitive psychology phenomenon much
developed in the 1980s that states that people judge a loss of a given amount as more painful than they
judge the gain of an equal amount – try to take a toy away from a toddler and you will quickly be
reminded that the angst of a loss is bigger than the joy of receipt! Applied to grownups and their
shopping behavior, prospect theory would imply that if I take up the price if your favorite gadget by 50 €,
your disappointment will be greater than the joy you would feel if I took the price down by 50 €..
5 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Figure 1.2 prospect theory as a price sensitivity curve
However, neither the symmetrical price elasticity belief (figure 1.1), nor academic prospect theory (figure
1.2) are right. When we look across studies, we see that shoppers are less sensitive to price increases
than to decreases. In fact, we see that on average, products are 30% less elastic to a price increase
than they are to a price decrease.
Putting this in perspective – reporting a 30% difference between up and down pricing trends at such a
macro level is a noteworthy effect. This is as big as the difference in price elasticities between Russia
and the United States (see belief 2), it is also on the same scale as the difference in sensitivities
between discount retailer shoppers vs. hi-low retailer shoppers (see belief 3).
Figure 1.3 The reality of CPG price elasticity curves
6 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
We believe this effect can be explained quite simply. CPG markets are open markets; they have many
brands, with multiple variants, perfumes, sizes, etc. Sales for any one of the products on shelf can be
broken down into sales coming from its loyal consumers, and sales coming from non-loyal consumers,
or switchers. Loyal consumers tend to be less price sensitive, where switchers tend to be more
sensitive.
The proportion of loyal to non-loyal consumers will depend on many factors, for example, the product
category, the product’s branding in that market, where the product sits in the brands’ portfolio
architecture, etc. Nonetheless, every product in every market can be simply thought of in terms of
demand coming from loyal consumers and switchers.
As you increase price, sales coming from switchers will drop; as you decrease price, sales coming from
switchers will increase. For the most part, sales from loyal consumers will remain relatively stable as
prices change. This means that as you increase your price, loyal consumers become relatively more
important in determining your product’s price elasticity as they are a bigger part of your user base; as
you decrease prices, the switchers determine a bigger part of your product’s price elasticity. The net
result is a more sensitive reaction to down pricing, than to up-pricing.
Figure 1.4 Explaining the reality of CPG price elasticity curves
Prospect theory and the symmetrical pricing belief are debunked as myths in a world of open market
where consumers have choice to change easily from one product to another as prices move. When
making pricing decisions, it is important to differentiate between the price elasticity for taking price down,
from that for taking price up [ref 2].
7 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Belief 2: The higher the standard of living in a country, the lower price sensitivity of
consumers.
A common assumption is that people are mainly price sensitive when economic context requires them to
be – said most simply, the better off we get as consumers the less price sensitive we become.
This assumption is sometimes criticized as naïve. Consumers have a strong sense of what is a fair price
to pay for goods which is grounded in references to past or competitive offerings, and will not be willing
to pay more just because their purse strings are able to stretch more.
Ironically, there are some people who would argue that we might see a reverse effect of consumer
affluence and price elasticities. Over the last 10 years, relatively affluent markets have shown increasing
consumer scepticism and proliferation of white label goods. The net effect is heightened price elasticities
as consumers scrutinize prices and seek better value for money. This thinking flies in the face of the
assumption that price sensitivities are related to the standard of living.
We wondered if our database could shed some light on this debate.
First we looked into the average price elasticities for any country, across categories, where we had
robust samples sizes and a fair spread of categories behind the data.
Then we looked for a simple measure of consumer affluence. The simplest measure of a nation’s wealth
is GDP (Gross Domestic Product), or approximated to the consumer level, the GDP per capita. We
looked at GDP per capita that has been adjusted for purchasing power parity (GDP PPP) to better
approximate the consumer affluence of a country. [ref 3]
UK
France
USA
Netherlands
Germany
Spain
Italy
Russia
Average SKU Price
Elasticity to Up Pricing
-1.0
-1.1
-1.1
-1.1
-1.1
-1.2
-1.3
-1.3
GDP/Cap at PPP ($)
34,619
33,679
46,381
39,938
34,212
29,689
29,109
14,920
Sample Size (#
SKUs)
936
328
143
233
705
344
225
246
Table 2.1 GDP PPP/Cap
Our data is based on eight countries, six of which are in Western Europe. Talking correlations is
therefore too ambitious. Nonetheless, we dare to give a more anecdotal take on what we see.
We see evidence that “extremes” of consumer affluence (moving from Russia to the United States) may
be related to price sensitivities. Consumers in the United States and Northern Europe consumers are
8 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
less price sensitive, where GDP PPP/cap rates are higher; Italian, Spanish and Russian consumers are
more price sensitive, where we see lower GDP PPP/cap rates.
Figure 2.1 shows this data mapped, along with an averaged European data point. There seems to be a
gross negative trend between consumer affluence and price sensitivity at a very macro level – on a
basis of three data points, granted.
Figure 2.1 Average price elasticity by GDP PPP per Cap
The variability between European countries, between more affluent European countries and the United
States, or between less affluent European countries and Russia cannot be explained by GDP PPP/cap
differences, certainly not at top level.
We conclude that major increases in national affluence seems to enable lower price sensitivities, but that
more minor differences such as those seen across different countries in Europe, do not determine price
sensitivities within a market. − Hence economic differences within Europe. Changes in price sensitivities
for a country are likely better explained by product innovation, commercial innovation, and changes in
retailing strategies.
Another way to understand how consumer affluence affects price sensitivity is to examine cases where a
category was measured before and after the 2009 economic recession. We have a number of examples
in Europe that map the price sensitivities of SKUs in a market measured in 2006-2008, and then during
2009 again. The following case represents well the general picture of what is typical.
Two studies were carried out for a client in Germany. It was for a category that could be considered as
moderate category involvement, and private label brands have played a big role in this market for at
least five years.
In early 2008 we undertook a market level sensitivity study, and in late 2009 we undertook again an
almost identical study.
9 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
What we saw here was that while GDP PPP/cap had dropped by almost 4% (IMF 2009 World Economic
Outlook), there was no change in overall SKU level sensitivities in this market
Total Market
Branded goods
only
2008: Av. SKU Price
Elasticity to Up
Pricing
Sample Size
(# SKUs)
2009: Av. SKU Price
Elasticity to Up
Pricing
Sample Size
(# SKUs)
-1.0
40
-1.1
26
-0.9
36
-0.9
20
Table 2.2 Price elasticity by type of brands
Product level elasticities changed over this time, but not at the overall market level. Changes in
elasticities for specific products made sense in light of the innovation and communication changes that
had happened 2008 and 2009. In other words, the successes of strong marketing campaigns had much
more influence on the price elasticities than what was happening in the economy at large.
Our data partially supports the belief that consumer affluence will lessen price sensitivities of consumers.
We believe that large scale differences in consumer affluence have the potential to change consumer’s
sensitivities. However we also believe that small changes in affluence between developed countries, or
within countries as markets boom and recess will have little influence on price elasticities; what does
have major influence on price elasticities is the marketing activities of the brands playing in the market.
10 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Belief 3: German consumers are very price sensitive; more price sensitive than
consumers in other European countries.
We simply do not see any evidence to support this belief.
UK
France
Netherlands
Germany
Spain
Italy
Av. SKU Price Elasticity to
Up Pricing
-1.0
-1.1
-1.1
-1.1
-1.2
-1.3
Sample Size (#
SKUs)
936
328
233
705
344
225
Table 3.1 Average price elasticity by country
Digging a little deeper, we wondered whether this conclusion is an artefact of our data – just maybe the
categories in our database for Germany are less price elastic than the categories studied in the United
Kingdom, for example.
Just to be sure, we looked at the category for which we have most data. Table 3.2 summarizes the price
elasticities across European countries when looking at one such category in isolation – a household
cleaning category with where retailer brands are strong in every country considered. Here we see that
Germany actually ranks as the least price elastic of all countries considered! Here again we see
absolutely no evidence to support the belief that German consumer are more price sensitive than their
European neighbors.
Spain
France
UK
Italy
Germany
Average SKU Price
Elasticity to Up Pricing
Rank 1 (most price elastic)
Rank 2
Rank 3
Rank 4
Rank 5 (least price elastic)
Sample Size
(# SKUs)
100 – 200
50-100
>200
50-100
50-100
Table 3.2 Ranked price elasticities across Europe for one major
household cleaning category
Now that we have clearly debunked this belief as myth, we turn our attention to understanding the origin
of this belief. Perhaps it is the association of Germany with Aldi and Lidl, two of Europe’s biggest
discount retailers, where the retail marketing message is guaranteed everyday low prices. It makes
sense that these channels are more price sensitive as they attract more thrifty consumers in the first
place. Our data support this idea as table 3.3 shows.
11 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Hi-Low Retailers Only
EDLP/Discounter Retailers
Average SKU Price Elasticity to
Up Pricing
-1.1
-1.5
Sample Size
(# SKUs)
50 - 100
50 – 100
Table 3.3 Average price elasticity by retail channel in Germany
Indeed, there is evidence that shoppers of discounter channels generally are 30% – 40% more sensitive
to price increases than shoppers of Hi-low retail channels.
In summary, we have shown that it is a myth that German consumers are more sensitive than other
Europeans. However, it is clear that discounter retailing strategies are associated with higher price
elasticities. This may be because they attract more price sensitive consumers to start, or because they
encourage price sensitive behavior.
12 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Belief 4: It is better to change your pack size than to change your shelf price if you
want to increase prices.
We compared average price and size elasticities in our database. Specifically, we compared how
shoppers react to an identical price change when executed via a shelf price change ($ 3.49 to $3.99 for
example) or a pack size change (74 ml. to 64 ml. for example).
What we found is that for price increases, the difference in sensitivity between playing with price vs.
playing with size is not significant. For price decreases, reducing your shelf price is a more efficient way
to gain volume.
This implies it is better to play with your price than with your pack size whether you want to price up, or
price down. When pricing up, both methods are equally effective to sustain volume sales, and when
pricing down you will increase volume sales more if you change your price rather than pack size.
Figure 4.1 Average price vs. average size elasticity
Given that changing pack size is a costly business, requiring changes in production lines, manufacturing
and package designs, this is great news for most people. Generally speaking we are saying you don’t
need to change your pack sizes, but change your price instead.
Having said this, we have seen many cases where there is clear exception to this general rule. Cases
where downsizing the pack is clearly a better option than taking price up. Most of these cases involve
price barriers, or strong category norms towards line pricing.
This belief is debunked as myth: overall, a size change is equally, or even less effective than a price
change. However, category norms on how pricing is presented to the shopper mean ad hoc research is
be needed to know what to do in your case.
13 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Belief 5: Women are more price sensitive than men.
It is often said that women are more price sensitive than men, simply because they are often in charge
of managing the household budget and are therefore more inclined to look for good deals and value for
money products. As men are more often buying products for themselves, they are thought to be less
price sensitive. But is that true? As we will see in belief 6, higher involvement does not necessarily lead
to lower price sensitivity.
In order to shed some light on this matter, we looked at the price elasticity of female dominated
categories vs. male dominated categories, where female dominated categories are defined as
categories in which the vast majority (>70%) of respondents in our pricing studies were female
consumers (and similarly for male dominated categories).
Female dominated categories
Male dominated categories
Average SKU Price
Elasticity to Up Pricing
-1.1
-1.0
Sample Size
(# SKUs)
2308
345
Table 5.1: Price elasticity by categories dominated by males/females
The results show that female dominated categories are slightly more price elastic than male dominated
categories. The gap is very small though and is not significant.
Our database only holds very few studies and categories where we have a large enough sample of both
men and women to be able to analyse price elasticity differences between men and women within these
categories. What we have seen in these few studies though is that there is no significant difference
between men and women.
Therefore, even though it looks like there is some truth in this belief looking only at the average price
elasticity of male vs. female dominated categories, we cannot fully support it.
14 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Belief 6: The lower the personal involvement of a category, the higher the price
elasticity.
This is something we hear quite often in the CPG industry. The assumption is that categories with high
personal involvement, products that are closer to consumer’s hearts, minds and skin, are more precious
to consumers, and therefore she/he won’t mind paying that little bit extra if needed.
One of the problems we face in addressing this belief is that there is no objective measure of personal
involvement of a category. Furthermore, brand managers working on most categories will be ready to
argue that theirs is a highly personally involved category.
Approaching the question from a different angle, we can take a look at the price elasticities to see if
there is a pattern between this and the category definitions. We have created four broad categories of
products as table 6.1 shows: 1. Household cleaning, 2. food, 3. personal beauty & hygiene and 4.
consumer health products.
Before going any further though, this table is unusual compared to all other tables in this paper; there is
quite some diversity in price elasticity indices. In exploring the seven beliefs of this paper, this is the one
cut of data where we see that there really are differences on how shoppers behave when asked to pay a
bit more than usual. But this doesn’t tell us yet whether the differences in price elasticities are driven by
consumers’ personal involvement with a category, or whether it is driven by other factors.
Looking at table 6.1 we see that food categories show a pretty even spread of price elasticities. While
you could argue that foods also span high to low category involvement we can tell you that even if the
food types were more transparent, there would be no clear pattern emerging between personal
involvement and price elasticity. For example, one of the most price elastic food categories in this list is
beer – surely a highly personally involved category for male consumers, and one of the least price
elastic food categories is a basic food ingredient – surely one of the least personally involved categories
most people would argue.
15 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Average SKU Price
Elasticity to Up Pricing
-2.1
-2.0
-1.7
-1.6
-1.4
-1.4
-1.3
-1.1
-1.1
-1.0
-1.1
-1.0
-1.0
-1.0
-1.0
-1.0
-1.0
-0.9
-0.9
-0.9
-0.9
-0.8
Consumer Health
-0.8
Food
-0.7
Food
-0.6
Table 6.1: Price Elasticities by Category
Category
Household Cleaning
Food
Household Cleaning
Household Cleaning
Food
Personal Beauty & Hygiene
Household Cleaning
Consumer Health
Personal Beauty & Hygiene
Food
Personal Beauty & Hygiene
Personal Beauty & Hygiene
Household Cleaning
Food
Household Cleaning
Personal Beauty & Hygiene
Personal Beauty & Hygiene
Food
Consumer Health
Food
Personal Beauty & Hygiene
Personal Beauty & Hygiene
Sample Size
(# SKUs)
16
84
133
21
253
171
19
23
92
39
129
38
63
47
601
169
134
107
201
52
53
169
75
142
35
Nonetheless there is a picture that emerging from this data: household cleaning products are
significantly more price elastic than personal beauty, consumer health and personal hygiene products. In
other words, consumers are more willing to stretch to paying more for products that are close to their
skin or used for taking care of themselves. Consumers are less willing to pay more for products being
used to clean their homes, offices, cars and clothes.
Finally, consumers show a broad range of willingness to pay more for foods. And much as we have
tried, we have never been able to find any patterns dictating which food products will be more or less
sensitive beyond the marketing strategies of the companies manufacturing and marketing them. We
have found a kernel of truth behind the belief that low personally involved categories show higher price
elasticities if we are willing to: 1) exclude all food categories from this statement 2) define personal
involvement as self-pampering, and not taking care of the home.
16 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Belief 7: Price elasticity is higher with larger packs than with smaller packs
It seems likely, but is this true? And is this always the case for expensive and inexpensive categories
alike?
As we look across categories, some have an almost random spread of pack sizes being offered to
consumers, and other have clear pack sizes offerings that all market players are offering in harmony:
small medium and large packs, for example. Examples of categories where there are clear pack size
offerings from all major players: diapers, beer, snack foods. For these categories, you can normally
divide packs in market into “small”, “medium” and “large” packs, where the “medium” pack size is more
or less the average size in the market.
And here we find evidence of a belief clearly based on fact. We see that the price elasticity of large pack
sizes is indeed higher than for medium and small sizes (table 7.1).
Why are larger packs more price elastic than smaller packs? We believe it is likely to be due to a couple
of reasons. Firstly, because larger packs tend to be bought by consumers who are more price-sensitive
to begin with as larger packs tend to be a better deal than smaller packs. Secondly, it can be due to the
”out of pocket” effect: a 10% price increase on a large pack is a much bigger price change than a 10%
price increase on a smaller pack. For example, a large pack of diapers could cost as much as 25 €,
where a small pack costs only 6 €. A 10% price increase results in only an additional 60 c. on the small
pack, but €2.50 on the large pack – a harder price increase to live with in absolute terms.
All Sizes
Small Packs
Medium Sized Packs
Large Packs
Average SKU Price
Elasticity to Up Pricing
-1.0
-0.9
-1.0
-1.2
Sample Size
(# SKUs)
1854
670
764
420
Table 7.1 Average price elasticity of different pack sizes
This belief is confirmed as true: large packs are more price elastic than small and medium packs.
17 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
Conclusion
Altlas
We’ve looked at seven great pricing beliefs and debunked many as myths with some simple but
compelling data.
One common emerging theme is that the most relevant trends for price elasticities are not driven by
natural consumer tendencies; rather the differences we see are better explained by the marketing and
retailer strategies going on in CPG markets.
For example, in belief 2 we looked at a case we studied where the 2009 economic recession had no
impact on market level price elasticities. On the other hand, despite this “severe” economic down-turn,
some players in the market had managed to reduce price elasticity through effective marketing, where
others had weakened their elasticities by not remaining competitive enough.
In belief 1 we saw a strange phenomenon. Where individual shoppers may be more sensitive to price
increases than price decreases (prospect theory), macro market trends show that price increases are
less elastic than price decreases. This, we have argued, is because of the large choice that is available
to shoppers –once again a factor driven by the manufacturers and retailers of these markets which overrides any natural consumer preferences.
Debunking myths helps solve common dilemmas. For example, should you put less in a pack, or take
your price up? Most of our clients ask themselves regularly this question. Belief 4 looked at this head on.
While we saw that this really does depend on the pricing norms in the category you are talking about,
18 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
generally speaking either way works equally well to sustain your volume as you take up prices.
Nonetheless by the time you factor in the capital costs of changing your packaging for downsizing and
consider the risk you are taking by driving down consumption by selling less product at every purchase,
a simpler price increase is probably a more attractive option in most cases.
Belief 7 was not debunked as myth, it is true that shoppers are more price sensitive when buyer large
packs than smaller packs. But this is because we, as shoppers, have been trained to expect better value
for money when shopping big packs. Once again, market norms now dictate how consumers will
behave.
The danger with myths is that if everyone in the industry believes them, we may end up making them a
reality. Take for example belief 5; we saw no evidence to support the belief that women are more price
sensitive than men. The problem is, if we start believing that based on this, women are more price
sensitive than men, and placing more emphasis on price differentiating marketing activities in female
focused categories (more special packs, more promotions, communicating price when talking about
value for money) we may actually train women in this category to become more price sensitive. And
making our consumer more price sensitive is the last thing any branded goods manufacturer should be
trying to achieve.
We hope these seven simple examples have illustrated the value of stepping back from individual pieces
of research to get a bigger picture. Debunking myths helps prevent missed marketing opportunities; next
time you hear “we can’t launch this in Neverland, consumers there are too price sensitive”, or “with the
current economic climate, now is not the time to launch a premium product” think again. Are your
assumptions well founded?
Beyond thinking about the seven pricing assumptions addressed in this paper, we hope we inspired to
think through marketing assumptions you are making every day. When working on a brand, “joining the
dots” means looking across different study types to gain brand specific insights; agencies too can join
the dots across studies to gain cross category and inter-regional insights. Are you, or is one of your
trusted agencies sitting on a goldmine of information that can be leveraged to drive smarter decisions
without needing ad hoc research every time?
19 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
References
1. Kahneman, Daniel, and Amos Tversky (1979) "Prospect Theory: An Analysis of Decision under Risk",
Econometrica, XLVII (1979), 263-291.
2. Chen, S-F.S., K.B. Monroe & Y-C Lou (1998). “The effects of framing price promotion messages on
consumers' perceptions and purchase intentions”, Journal of Retailing
3. Data Source for GDP/Cap is Data source is IMF 2009
The authors
Maureen Arink is responsible for Method Development, Consumer Research Division, SKIM,
Switzerland
Vicky Nef is Account Director and Division Manager, Consumer Research Division, SKIM, Switzerland.
Anne Favrelle, Consumer Research Division, SKIM, Switzerland.
Appendix
1. Methodology with which the data was collected: Choice-Based Conjoint analysis Price elasticity
(PE) can be measured using different methodologies. When looking at a measure that represents the
market and interactions between competitors following price, size or other changes in products, conjoint
analysis, and in particular Choice-Based Conjoint (CBC) is a good method to measure PE. Another
widely used type of method is econometric analysis based on sales data. However, this as this is limited
to historic events, and often at SKIM we need to measure future events, we use CBC in our pricing
studies.
In CBC, products are constructed of different product attributes like brand, price, pack size and
promotion. In order to estimate the sensitivity of consumers towards each of these different product
attributes, each consumer will be shown so-called choice tasks. In these choice tasks, they will see
shelves with products from which they select the product they would buy in that situation. The products
on the shelves are randomly generated by combining the different product attributes. Each consumer will
go through several choice tasks, each of which consists of a new set of products, with differing prices,
pack sizes and sometimes on-shelf price promotions.
While making the choice for the product they would buy in each choice task, the consumers reveal their
sensitivities towards the underlying attributes (brand, pack size, price, on-shelf price promotion). So for
instance, we will be able to see whether they consistently go for the cheapest product (then they are
highly price sensitive), whether they consistently buy the same product (highly brand / product loyal) or
whether they are very sensitive to promotions. Based on these sensitivities, which are expressed in
‘utility values’, we will be able to predict the choice of each consumer in different market scenarios. By
combining the results for the different consumers into a simulation model, we can run different ‘what if’
scenarios which then show the effect of e.g. price changes on volume shares.
20 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com
2. Database: size, measures included, number of SKUs, categories, countries
The results of price changes for each product in the test, for instance the effect of a 10% or 20% price
increase or decrease on the volume share of these products, the effect of different promotions on
volume share, are brought together with other studies in a database.
This database contains is currently based on 108 pricing studies, conducted in 16 countries and 38
categories. This results in a database with price elasticity information on 3,633 products.
The information in the database includes, amongst others:
• category
• country
• channel
• base volume share
• base price
• base pack size
• % male / female in sample
• Effect on volume share of price change (-20%, -15%, -10%, -5%, +5%, +10%, +15%, +20% price
change)
• Effect on volume share of size changes (-10%, +10%)
• Effect on volume share of different promotion types
• Price awareness
• Brand awareness / stated loyalty
This database enabled us to look at price elasticity by different countries, categories, pack sizes etc. and
was used in finding the answers to the pricing assumptions discussed in this paper.
21 / 21
Rotterdam | Geneva | London | New York
www.skimgroup.com