Modeling the Impact of Marketing Mix on the Diffusion of Innovation in the Generalized Bass Model of Firm Demand


 

Page 75 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

ABSTRACT 
 

In the field of business simulation and gaming there has 

been much attention given to the modeling of demand, but 

very limited research on the role of the marketing mix for 

new and innovative products. This study examines how the 

marketing mix impacts the diffusion of demand for new and 

innovative products using the framework developed by 

Frank Bass (1969). The Bass Model was chosen because it 

is widely accepted within academia as a structured and 

disciplined approach for forecasting the intertemporal dif-

fusion of demand. Our concern with the Bass Model is the 

assumption that market potential is fixed and independent 

of the marketing mix. Based on a careful review of the liter-

ature, revisions to the model were derived and tested by 

running twenty-eight computer simulations with changes in 

the pre-specified levels of both price and advertising, given 

varying assumptions with respect to the elasticity coeffi-

cients in the marketing mix function. Our research suggests 

that management of the marketing mix effects all the key 

elements of the Bass Model including the coefficients of 

innovation and imitation, the total market size and thereby 

the rate of diffusion. These findings are significant with 

respect to the forecasting of demand and the effective de-

sign and internal validity of business simulations used for 

management education. 

 

INTRODUCTION 
 

The diffusion of innovations into a market and the role 

of the marketing mix has been a topic of scholarly research 

in multiple disciplines, including but not limited to market-

ing, economics, operations management, statistics, and 

technology management.  The topic is important to both 

academics and practitioners because the expected pattern of 

diffusion impacts a firm’s decisions about its supply chain, 

manufacturing, inventory, and marketing investments. 

In the field of business simulation and gaming there 

has been much attention given to the modeling of demand, 

but very limited research on the role of the marketing mix 

for new and innovative products despite the fact that fifty 

percent of business profits are from products that are less 

than five years old (Teach and Schwartz, 2003).  Goosen 

(2010) notes that most business simulations have demand 

algorithms which are based on classical economic theory, 

which does not address the issue of the intertemporal diffu-

sion of new and innovative products.  With respect to the 

important role of advertising in the marketing mix, in the 

37 years of ABSEL history (an academic association focus-

ing on business simulations) only a few papers can be 

found that deal directly with the algorithmic modeling of 

advertising as a demand variable (Goosen, 2010).  The re-

cent studies by Goosen (2010, 2011) and several past stud-

ies explore the impacts of the marketing mix and advertis-

ing on firm demand, including papers by Gold and Pray 

(1983, 1998), Goosen (1986), Carvalho (1991), Teach 

(1990), Thavikulwat (1988) and Cannon (1994, 1996), but 

do not deal with the complexities of diffusion of new and 

innovative products. 

The word “diffusion” originally was applied to the 

acceptance of a technology-based innovation within a so-

cial group over time (Brown, 1991, Stoneman, 2002).  This 

“technology insertion” perspective on diffusion manifested 

itself in the influential theories of Everett Rogers described 

in the Rogers 5 Factors framework (Rogers 1995) and more 

recently in the concept of chasm crossing developed by 

Geoffrey Moore (Moore, 1991, revised 1999).   

Because of the wide acceptance of the idea of the dif-

fusion of technological innovations as a social phenomenon 

the concept has been adopted by strategists and marketers 

in a variety of non-technical product categories.  For exam-

ple the social nature of diffusion is now being applied to 

the entertainment industry (movies, TV shows, music, con-

certs, plays), video games, restaurants (through on-line 

reviews, blogs, and tweets), fashion (think of Crocs), and 

wines (South America and Spain versus “old country” 

wines). To be most relevant, the product’s diffusion must 

be influenced by the communication within a social system 

of an event or events which are the proximate cause for an 

increased product demand.  The event or events themselves 

need not be innovative; and the word “product” is used in 

MODELING THE IMPACT OF MARKETING MIX ON THE DIFFUSION OF 

INNOVATION IN THE GENERALIZED BASS MODEL OF FIRM DEMAND 
 

Robert Boehner 

Rochester Institute of Technology 

rboehner@saunders.rit.edu 

 

Steven Gold 

Rochester Institute of Technology 

stevengold@saunders.rit.edu 

mailto:rboehner@saunders.rit.edu
mailto:stevengold@saunders.rit.edu


 

Page 76 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

its most expansive meaning.  “Product” refers to a physical 

good, service, idea, or even a person (Madonna, Lady Ga-

ga, and “the Donald” all are marketed as products).  The 

product need not be technology based or innovative.   

 

PURPOSE AND PROCEDURE 
 

The purpose of this study is to examine and critique 

how diffusion impacts demand for new and innovative 

products using the framework developed by Frank Bass 

(1969). The Bass Model was chosen because it is widely 

accepted within academia as a structured and disciplined 

approach to forecasting new product demand and the diffu-

sion of innovation.  In addition the Bass Model has gained 

wide acceptance by practitioners as a forecasting methodol-

ogy (Gueso and Guidolin, 2009).  

The paper is composed of five major sections and will 

proceed as follows:  

 Describe the Bass Model and one of its extensions 

referred to as the Generalized Bass Model. We will 

discuss the strengths and weaknesses of the model, and 

make recommendations for change.  The recommenda-

tions will focus on the way the marketing mix ele-

ments affect the total market size in the Bass model.  It 

is our hypothesis that the market size is dependent on 

the marketing mix, necessitating an important change 

in the structure of the Bass model. 

 Review the literature on the impacts of the marketing 

mix on market size, focusing on two key elements, 

price and promotion. The review is segmented into two 

parts. The first part examines the literature from a 

practitioner’s perspective; and the second part from an 

academic perspective. We find high consistency be-

tween these two perspectives and much support for our 

hypothesis. 

 Develop a revised Bass Model, making the market size 

a function of the marketing mix. The revised model is 

illustrated with a numerical example, assuming param-

eters estimated for the adoption of optical scanning 

equipment. The example serves as a standard of com-

parison to test the behavioral impacts of changes in the 

marketing mix on diffusion of new and innovative 

products.  

 Test the revised Bass Model by developing a set of 

examples and simulating the market demand. In this 

study twenty-eight computer simulations are tested 

with changes in the levels of both price and advertis-

ing, given varying assumptions with respect to the 

elasticity coefficients in the marketing mix function. 

Figures illustrating the results of each of the major 

simulations are shown and explained. 

 Conclude by highlighting our findings and the signifi-

cance of the behavioral changes in the revised Bass 

Model and implications with respect to effective busi-

ness strategies and the design of business simulations.  
 

BASS MODEL BACKGROUND 
 

In 1969 an analytical model for forecasting the first 

purchase of a new product category was published by 

Frank Bass in the journal Management Science (Bass, 

1969).  Bass’ model is now described as a diffusion model 

because its hypothesis posits that the growth in demand for 

a new product category is a function of how information 

about that product is diffused in a social system.  Bass’ 

work is now referred to as “The Bass Model.” (Elie Ofek, 

2005).  In 2004 Bass’ original article was selected as one of 

the ten most frequently cited papers in the 50 year history 

of Management Science (Management Science 15[5] 2004).  

It was the only marketing related article on the list. 

The significant contribution of Bass is his assumption 

that the probability of additional first-time adoptions of a 

new product in a future time is the function of the number 

of consumers who have already adopted the product.  Thus 

Bass considers new product diffusion to be viral, essential-

ly stating that diffusion occurs in a way similar to the 

spread of a viral disease in society.   

The model has been widely accepted.  Over 200 aca-

demic articles have been published about the application of 

the model.   The model was initially applied by Bass to 

forecast the demand curve for durable goods, a product 

such as machine tools, cars, or home appliances that have a 

long life-cycle and that are purchased infrequently by con-

sumers and businesses.  Over the past 20 years businesses 

have employed the model to forecast adoption of a wide 

variety of products, including satellite TV, satellite radio, 

refrigerators, calculators, CD players, home PC’s, and cell 

telephones.  The Bass Model has also been extended to 

forecasting demand for individual brands and niche prod-

ucts (Krishman and Bass, 1994).  Norton and Bass (1992) 

applied the model to a wide range of non-durable product 

classes including pharmaceuticals and recording media.  

Jain et.al. examined how supply constraints can be integrat-

ed into the model.  Companies such as Kodak, RCA, IBM, 

Sears, and AT&T have used the Model internal to their 

businesses (Mahajan et.al., 1990). 

Application of the Bass Model creates a curve that 

specifies the period and cumulative first-time adoptions of 

a new product category.  The model itself is relatively sim-

ple.   Bass identifies two types of new product adopters: 

 Innovators: Adopt new products independent of the 

actions of others within a social system.  They adopt 

through “internal influence” (Vijay Mahajan et.al., 

1995).  In the model their adoption rate is represented 

by the letter p.  The value of p is referred to as “the 

coefficient of innovation.” 

 Imitators:  Their adoption of new products is influ-

enced by the adoption rate of others in a social system.  

They respond to input from others, specifically the 

communications of those who have already adopted 

the product.  In the model their adoption rate is repre-

sented by the letter q.  The value of q is referred to as 



 

Page 77 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

“the coefficient of imitation.” 

Bass states that the likelihood of additional first-time adop-

tions of a new product by imitators in a future time is a 

linear function of the number of previous adopters, with the 

actual number of adopters limited by the total market size, 

represented by m in the model.  In other words previous 

adopters influence the adoption rate of new adopters in the 

current time period (for example, this month) and in future 

time period (for example, for the rest of this year). 

 

THE MODEL ITSELF 

 

As mentioned above the output of the Bass Model is a 

curve that represents period and cumulative adoptions of 

new products by both innovators and imitators.  In the 

model: 

m = the total market (expressed as unit sales) for a 

product category 
p = the probability of adoption by innovators (referred 

to as the “coefficient of innovation”) 
q = the probably of adoption by imitators (referred to 

as the “coefficient of imitation”).  Because q is 

influenced by the number of previous adopters 

within a social system q is multiplied by the num-

ber of consumers who have already adopted the 

product at the start of time period t.  This calcula-

tion captures the idea at the heart of the Bass 

Model: The probability of adoption by a new imi-

tator is a function of the number of consumers 

who have already adopted. 

The above leads to the following equation: 

The probability of adoption by a new imitator in 

time period t = p + (q/m)N(t-1)   

Where: 

N(t-1) = the cumulative adoptions from the prior time 

period 

 

An important and constraining assumption in the Bass 

Model is that the total market size (m) is a constant.  The 

market size is not impacted by externalities or by the ac-

tions of participants in the industry, such as significant 

changes pricing, advertising, or other elements of the mar-

keting mix. 

 

EXTENSIONS TO THE BASS MODEL 

 

The most important extension of the Bass Model is 

referred to as the Generalized Bass Model (Bass et.al 

1994).  The Generalized Bass Model examines the impact 

of marketing mix variables such as pricing and increased 

advertising and promotion on demand levels.  Within the 

Generalized Bass Model changing these marketing mix 

elements shifts the demand curve in time, e.g. the shape of 

the demand curve is changed, but total demand is un-

changed.  Importantly changing pricing and advertising 

have no impact on market size: m remains as a constant.   

 

Strengths and weaknesses of the Bass Model 

 

The most important strength of the model is that it 

seems to work.  It has accurately forecast the rate of diffu-

sion for a large number of product categories over multiple 

decades.   

One of the best examples of the forecasting accuracy 

of the Bass Model is a presentation given in 2001 by Frank 

Bass and others (Bass et. al. 2001).  In 1992 Bass was hired 

as consultant by DirecTV and asked to use his model to 

forecast the diffusion of the satellite TV product category.  

DirecTV had previously completed a stated intentions mar-

ket research study where consumers were asked about the 

probability of purchasing satellite TV.  A prior study by 

Jamieson and Bass (1999) allowed Bass and his colleagues 

to estimate m from the results of the stated intentions study.  

Bass “guessed by analogy” (to use his own words from his 

slides) on the value of p and q, extrapolating from the actu-

al diffusion rate of similar previously introduced products.  

The forecast of satellite TV adoptions developed by Bass 

was stunningly accurate.  Bass estimated a first year pene-

tration of satellite TV of 1.37% of US households; actual 

adoptions were 1.21%.  Bass’ 1992 forecast of cumulative 

demand from 1994 through 1999 was also accurate.  Bass 

forecast 9.4 million cumulative adoptions by 1999; actual 

adoptions were 10 million. 

Another example of the Bass Model’s relative accura-

cy is the forecast of the adoptions of satellite radio in cars.  

In 2005 Elie Ofek published an article that reviews multiple 

approaches to forecasting satellite radio adoptions.  The 

forecasts were done between 1997 and 1999, when the 

product category first emerged.  Subject area experts from 

Warburg Dillon Read, Donaldson Lufken Jenrette, and CE 

Underberg developed forecast using their own proprietary 

methodologies.  Ofek’s forecast using the Bass Model, de-

riving values for p, q, and m by triangulating from the actu-

al diffusion curve of what Ofek concluded were products 

similar to satellite radio.  In 2005 actual cumulative sales of 

satellite radio were approximately 3 million units.  Ofek’s 

Bass Model forecast was for about 7 million; Unterberg 

forecast 22 million, Donaldson Lufken 35 million, and 

Warburg Dillon 42 million.  Ofek’s forecast was off by 

over 100%, but it was much closer to actual demand than 

the other three forecast methodologies.  

Obviously the accuracy of the Bass Model is only as 

good as the assumptions for the model’s three critical varia-

bles: the coefficients of innovation and imitation (p and q), 

and an estimate of the total market size (m).  Bass et. al. 

(2001) recommend arriving at values for p and q by 

“guessing by analogy” accomplished by a review of actual 

p and q values from previously introduced products.  Esti-

mating m is more problematic.  One potential source for m 

is “intent to purchase” studies, which have been shown to 

be accurate for non-durable consumer products, but less 



 

Page 78 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

accurate for consumer and business durables. Tigert and 

Farivan (1981) compared Bass Model forecasts  to actual 

demand for quarterly and annual sales of optical scanning 

equipment for supermarkets in the United States.  They 

concluded that the model failed to predict sales levels cor-

rectly.  They acknowledged that the model “forces a disci-

plined approach to estimating market potential”, but they 

cautioned that “in-depth analysis of product/market struc-

ture is a mandatory requirement in order to justify the 

choices made regarding…initial estimates of market poten-

tial.”  They observed that key data related to this in-depth 

market analysis must include a macro-economic analysis, 

actual price trends, and marketing efforts including sales 

force and advertising spending. 

The Tigert and Farivan study mentioned above identi-

fies the major weakness of the model: the assumption that 

the market size is a constant.  The Model assumes that nei-

ther exogenous nor endogenous factors will impact the total 

market opportunity for a product.  Thus the total market 

demand for the product category is determined at the time 

of its introduction and does not change over the life of the 

product.  Mahajan et.al. (1990) indicate that “there is no 

rationale for a static potential adopter population.  Instead, 

a potential adopter population continuously in flux is to be 

expected.”   

Mahajan et. al. discuss the implications of the assump-

tion of a fixed market size.  The number of buyers in a 

product category is not increased by any of the following 

actions: 

 

 Effective management of the marketing mix by 

market participants, such as introducing new prod-

ucts, lowering prices, increasing advertising, or 

expanding distribution. 

 An expansion in the size of the social system with-

in which buying decisions are made (as is the case 

with modern social media such as blogs, tweets, 

and Facebook postings). 

 The rate of innovation for a product category such 

as has recently been experienced in the smart 

phone industry 
 

Related to the above, the intensity of competition in an 

industry, as is currently being experienced with the actions 

of Apple with its iPhone and Google with the Android op-

erating system where the firms themselves and their part-

ners have aggressively managed all aspects of the market-

ing mix. 
Accepting that market size is a constant has important 

implications for marketing strategy.  The Bass Model as-

sumes a market that is a zero sum game.  The only way a 

firm can increase demand is by taking market share from a 

competitor. If correct a completely static market would 

result in hyper-competitive “red ocean” (Kim and 

Mauborgne, 2005) markets, lowering prices and margins of 

market participants.  Firms would have less incentive to 

invest in innovations or increased capacity, creating market 

stagnation in the product category.   

 

Making “m” a Variable 

 

The Generalized Bass Model is the starting point for 

our analysis of the impact of the marketing mix on period 

and cumulative demand within the model.  As mentioned 

above our concern with the Generalized Bass Model is the 

assumption that total market size, expressed as m in the 

model, is fixed and can not be impacted by the marketing 

mix.  We propose changing the market potential, m, in the 

model from exogenous to endogenous, and making m a 

function of the effective execution of the marketing mix.   

In this paper we limit our examination of the impact of 

the marketing mix on total market size to pricing and pro-

motion (advertising, sales promotion, public relations, etc.).  

Pricing and promotion are analyzed independently.  We 

limit our scope to pricing and promotion for two primary 

reasons:   

 To reduce the complexity of our analysis by isolating 

their separate effects on market size; and  

 In recognition that pricing and promotion are the mar-

ket mix elements most frequently manipulated by mar-

keters in an effort to increase demand for their as-

signed products.  The other two P’s in the marketing 

mix, place (channels) and product, are important long-

er term demand creation strategies.  Both have rela-

tively long lead-times – it often takes years to develop 

new products or open new channels. Price and promo-

tion are employed by marketers as both strategic and 

tactical marketing actions.  We believe that product 

and brand managers will be particularly interested in 

the impact of price and promotion on total market size, 

and hence on product category demand. 

 

LITERATURE REVIEW OF MARKET SIZE 

– PRACTIONER’S PERSPECIVE 
 

One of the authors of this article was a marketing and 

strategy practitioner for a Fortune 100 firm for 30 years, 15 

of them at the marketing and strategy vice presidential lev-

el.  He observed first hand the growth of the global copying 

and printing markets during the decades of the 70’s, 80’s, 

and 90’s.  From personal experience he observed that the 

actions of major market participants, specifically multiple 

frequent new product introductions, aggressive pricing ac-

tion, product and brand focused advertising, and expansion 

of distribution channels separately and in aggregate results 

in significant product category growth. 

Anecdotal evidence from different industries suggests 

that maintaining the market size as a constant in the Gener-

alized Bass Model does not match actual experience.  

There are multiple examples of changes in the marketing 

mix increasing the size of the total available market.  For 

example: 



 

Page 79 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

 When AOL introduced its unlimited Internet us-

age for $19.95 pricing plan the number of adopters 

of the service increased dramatically. 

 When the price of first desktop PC’s and then lap-

top PC’s broke through the $1000 price point de-

mand increased substantially. 

 Advertising for on-line brokerage services such as 

e*Trade, ScotTrade, and TD AmeriTrade in-

creased the number of brokerage accounts. 

 The combination of advertising and Internet-based 

social network spread the word about the inde-

pendent movie My Big Fat Greek Wedding and 

turned it into one of the highest grossing movies 

of that year. 

 The government limited cigarette advertising be-

cause of multiple research studies (cite) indicating 

that cigarette advertising increased the number of 

new users and reinforced the behavior of existing 

smokers, a clear indication of the impact of pro-

motion on total market size. 

 Advertising by pharmaceutical companies appears 

to create a market for new drugs that address med-

ical problems that many consumers probably did 

not know they have, such as Peripheral Artery 

Disease (PAD), and toe nail fungus.  Advertising 

was instrumental in creating and growing the mar-

ket for ED drugs such as Viagra and Cialis. 
 

Recently Apple’s pricing and advertising for the iPad 

created the market for tablet computers – a market that did 

not exist as a mass market before Apple’s success. 
Geoffrey Moore’s book, Crossing the Chasm, Market-

ing and Selling High Tech Products to Mainstream Cus-

tomers (1991, revised 1999) has had an important impact 

on the thinking of practitioners about how marketing strate-

gies impact the total market opportunity for a product cate-

gory.  Troy Byers of Stanford Technology Ventures Pro-

gram commented in 2006 that the book is “still the Bible 

for entrepreneurial marketing 15 years later” (Byers, 2006).  

As the title suggests Moore’s book emphasizes marketing 

high technology products.  But its concepts – particularly 

the idea of the “chasm” itself and the need for marketers to 

create the “tornado” to cross the chasm have been adopted 

widely in a wide variety of industries.  For example, pre-

viewing summer blockbuster movies on the Super Bowl 

which is broadcast in early February, is an example of cre-

ating the “buzz” that starts the “tornado” that starts the film 

“across the chasm”.   

Moore borrows from the theories of Everett Rogers 

(1988) in developing the concept of chasm crossing.  Rog-

ers identified five factors that influence the diffusion of 

innovations: superior product performance compared to 

alternative methods, compatibility with the customer’s val-

ue set (e.g. “this is the way things should be” – for example 

expensive wines should have corks, not screw tops), lack of 

complexity, convenient trialability, and easy observability 

(e.g. non-users can observe others using the product).  The 

willingness of early adopters of iPhone and iPad to demon-

strate their product to their friends is a good example of 

both trialability and observability. 

Rogers’ concept of trialability and observability are 

similar to Bass’ theory of the diffusion of innovations as a 

social phenomenon.  Trialability and to a even greater de-

gree observability depend upon prior adopters sharing their 

user experience with non-adopters.  Moore adds to Bass’ 

and Rogers’ thinking by arguing that there is a chasm be-

tween early adopters of a product (whom Moore calls tech-

nology enthusiasts and visionaries) and the early majority 

pragmatists.  The early adopters behave like Bass’ innova-

tors; the early majority like Bass’ imitators.  As Bass before 

him Moore states that these customer segments, the early 

adopters and early majority, have different buying motives 

and different buying behaviors. 

Moore says that specific marketing strategies are re-

quired to cross the chasm between the early adopters and 

early majority.  Much of Moore’s recommendations are 

focused on the need for identifying a specific target market 

and developing the “whole product” that satisfies the needs 

of the identified target market.  The whole product must be 

supported by a focused marketing strategy, including ad-

vertising, promotion, public relations, and viral marketing 

to build “buzz” about the product, and also supported by 

appropriate pricing and distribution strategy.    

Moore clearly believes that by following his prescrip-

tion for crossing the chasm a firm can increase the market 

opportunity for its products, that is, increase the size of the 

market.  Moore identifies five customer segments: innova-

tors, early adopters, early majority, late majority, and lag-

gards.  He suggests that success with one customer segment 

(early majority for example) depends upon success with the 

prior segments (first innovators and then early adopters, for 

example).  Because the buying behaviors for each segment 

are different Moore says that marketers should focus on 

one customer segment at a time, tailoring marketing mes-

sages and media to appeal to that group of customers.  

Crossing the chasm creates market momentum for the dif-

fusion of the product, expanding the market opportunity 

one customer segment at a time and continuously increas-

ing market size. 

Malcolm Gladwell’s popular book The Tipping Point 

(Little Brown, 2000) builds on the work of Rogers, Bass, 

and Moore by using the analogy of a disease virus to ex-

plain the diffusion of products and ideas into a market.  

Gladwell compares the diffusion to the rapid spread of viral 

epidemics.  He refers to ideas as “infections”; new fashions 

are “outbreaks”; new products are “viruses.  The role of 

advertising is to infect new consumers and build momen-

tum for the epidemic of mass diffusion.  A product, a trend, 

or an idea reaches the “tipping point” when a critical mass 

of adopters catches the infection.  The “tipping point” is 

somewhat analogous to crossing Moore’s “chasm” – ideas 

“tip” when the epidemic crosses a threshold and mass in-



 

Page 80 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

fections occur.   

Gladwell uses a different vocabulary to explain ideas 

already discussed in this article.  The “Law of the Few” 

directly relates to Moore’s early adopters; “The Stickiness 

Factor” is a build on Roger’s need for relative advantage, 

“The Power of Context” relates to Bass’ idea of diffusion 

as a social phenomenon and Roger’s concept of compatibil-

ity.   

In the “Law of the Few” Gladwell states that epidem-

ics are initiated by a small number of socially active people 

who transmit their infection to others.  He calls these peo-

ple “connectors, mavens, and salespeople.”  Connectors 

network effectively, mavens are recognized subject area 

experts, salespeople are articulate and persuasive champi-

ons.  All three are well connected, speak well and speak 

often, and start the contagion that spreads the virus.   

Gladwell emphasizes the importance of the “stickiness 

factor”, which is borrowed in part from Everett Roger’s 

“relative advantage”.  Products and ideas will not become 

viral unless they are compelling to a broad audience be-

yond the initial enthusiasts.  Getting through “the few” re-

quires winning over the connectors, mavens, and salespeo-

ple.  But if the idea or product doesn’t offer significant rel-

ative advantage the lack of enthusiasm of “the many” will 

cause the virus to die – it will not “tip”.  Gladwell says that 

stickiness is difficult to define and therefore is more com-

plex than Roger’s relative advantage.  Its presence depends 

on “context” (discussed below) and it may happen in an 

unexpected way that is contrary to conventional wisdom. 

Gladwell’s most creative contribution to the concept of 

the diffusion of product and ideas is what he refers to as 

“The Power of Context.”  Context strongly influences 

whether a product or idea will “tip” into general ac-

ceptance.  “Context” encompasses broad environmental, 

historical, and social factors that influence whether an idea 

is embraced by the mass audience and thus reaches the tip-

ping point.  In some ways Gladwell’s “context” is similar 

to Roger’s “compatibility” – is this idea or product con-

sistent with “the way things should be?”  Gladwell uses a 

broken window analogy to illustrate context, suggesting 

that when touring a neighborhood if a person sees broken 

windows that person will conclude that crime is a problem 

in that neighborhood and that property owners do not have 

pride of ownership.  Products and ideas will not “tip” un-

less they are consistent with the general public’s judgment 

that the context is right for the acceptance of the product or 

idea.   

The Tipping Point message for marketers is that diffu-

sion of ideas and products follows a predictable pattern.  It 

is essential to attract the endorsement of “the few.”  As 

Roger’s states, relative advantage is essential to 

“stickiness.”  Product endorsed by the few will not “tip” 

unless they have stickiness.  “Context” means that the idea 

or product is compatible with society’s broader attitudes; 

the idea of context is similar to Roger’s compatibility and 

reinforces Bass’ concept of diffusion as a social phenome-

non.  Gladwell states that marketers must understand and, 

to the degree possible, control context.  This can be done 

by creating strong emotional connections between consum-

ers and products, such as is done with effective branding 

campaigns.   

When marketers attract the endorsement of the few, 

when the product has stickiness, when the broader societal 

context is willing to accept the product, the market oppor-

tunity for the product will grow.  Thus, m will no longer be 

a constant; market size will increase. 

It is not the purpose of this paper to critique the con-

cepts of chasm crossing or the tipping point.  We observe 

that both books reach the same conclusion: diffusion of 

products and ideas follows a pattern, and when that pattern 

is understood and effectively managed diffusion will be 

successful, the market will develop stage-by-stage, and 

market size will increase.  The books reinforce the intuitive 

belief of marketers that effective implementation of fo-

cused marketing strategies will increase the size of the mar-

ket for their products.  In other words, the belief that mar-

ket size is not a constant.   

 

LITERATURE REVIEW OF MARKET SIZE 

- ACADEMIC PERSPECTIVE 

  
As mentioned earlier diffusion of innovations into a 

social system generally, and the Bass Model specifically, 

have been a popular topic for academic researchers.  Maha-

jan et. Al. (2009) includes a bibliography of 145 articles on 

this topic. Multiple academic articles identify potential 

problems with and attempt to improve on the assumptions 

of the Bass Model (Mahajan et.al 1990, Nguyen and Shi, 

2006).  Many of these articles identify Bass’ assumption of 

a static market size as potential problem area. 

The hypothesis of this paper is that industry partici-

pants do affect market potential through the management of 

their marketing mix.  Our specific focus, as mentioned 

above, is on the impact of price and promotion on market 

size.   

 

THE POLAROID VS. KODAK LEGAL CASE 

 

Perhaps the most thorough analysis of the affect of the 

marketing mix on market size is the examination of Koda-

k’s entry into the instant photography market that was then 

completely dominated by Polaroid.  (Mahajan et.al. 1993).  

Immediately after Kodak’s 1976 entry into the instant pho-

tograph market Polaroid sued for patent infringement.  In 

October 1990 a federal judge ruled for Polaroid and or-

dered Kodak to pay $909.5 million to Polaroid as compen-

sation for patent infringement.  Testimony and documents 

presented at the trial provided detailed information about 

Polaroid’s sales of instant cameras prior to Kodak’s entry, 

and the combined sales of Polaroid and Kodak instant cam-

eras when both competed in the market.  From a legal per-

spective the market data was important to establish sales 

and profits lost by Polaroid because of Kodak’s infringe-



 

Page 81 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

ment.  Polaroid completed detailed research to establish 

what the market size would be if Polaroid maintained its 

monopoly on instant cameras, how many potential Polaroid 

sales were diverted to Kodak cameras (providing the basis 

for calculating actual damage to Polaroid), and how many 

of Kodak’s camera sales were incremental (e.g. sales Polar-

oid would not have achieved, and thus an increase in the 

size of the market).  This case related research provided a 

rich trove of data for analysis.   

Kodak argued that because of its well established 

brand, the advantages of its instant camera products, and its 

well executed positioning and marketing strategy it created 

significant incremental demand for instant cameras, and 

therefore damages to Polaroid should be minimal.  In fact 

following Kodak’s entry sales of instant cameras increased 

substantially, from 3 million units in 1975 to more than 8 

million units in 1978.   

The authors used established models of diffusion of 

innovation to forecast, based on Polaroid’s actual unit sales 

prior to 1976, how many cameras it would have sold if Ko-

dak had not entered the market.  The analysis was compli-

cated by the facts that both Polaroid and Kodak had multi-

ple models of their cameras in the market, with different 

features and at different price points.  The analysis con-

cluded that: 

“More than two-thirds of the sales of Kodak instant 

cameras would not have been made by Polaroid if Ko-

dak had not entered the market.  Kodak’s entry ex-

panded the US instant camera market by 11.095 mil-

lion units, which it drew from its own pool of prospec-

tive buyers….  This implies that if Kodak had not en-

tered the total US instant camera market demand dur-

ing 1977 through 1985 would have been 32.635 units.  

Kodak’s entry expanded the market by 12.055/32.635 

(million units) = 37%, with most of this expansion tak-

ing place in the early years of the period.” 

Thus the market size for instant camera was not a con-

stant when Polaroid first created the product category.  Ko-

dak’s entry increased the size the market by 37%.  The au-

thors conclude that: “The addition of a new competitor may 

expand total market volume because market entry is usual-

ly accompanied by increases in product variety, promotion-

al activity, and distribution, as well as the reduction in 

price.”   Obviously the sentence above refers to the classic 

“4 P’s” of marketing: product, price, promotion, and place 

(distribution channels).  Because of the detailed demand 

information available the Polaroid/Kodak example presents 

a compelling argument for the effect of the combination of 

all four marketing mix elements (the 4P’s) increasing the 

market potential for a product category. 

 

PERSPECTIVES ON THE MARKET SIZE 

 

To understand how the marketing mix affects consum-

er buying behavior and overall market opportunity requires 

that we examine buyer behavior.  The core assumption of 

the Bass Model is that endorsement of a new product cate-

gory by members of a social system results in increased 

unit sales, e.g. the probability of a purchase by a non-user 

of the product is a function or prior users.  Assuming that 

Bass’ hypothesis is correct, it raises the question of how the 

social system influences non-users to purchase the product.  

Answering this requires an examination the psychology of 

buyer behavior. 

Buyers are risk averse (Gourville, 2006).  Buyers’ pur-

chase behavior is impacted primarily by two factors: in-

creasing awareness of the details of the new product, and 

the perceived price/value relationship (Kalish, 1985).  The 

more the consumer knows about the performance of the 

product (for example: features, reliability, warranties for 

many durable goods; entertainment value, food taste, ambi-

ence for many intangible products) the lower the buyer’s 

Figure 1A 

Standard Case of New Adopters per quarter with Market Size fixed 



 

Page 82 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

uncertainty about the quality of the product and/or the ex-

perience.  Information about the product is available from 

two sources: (1) actual users of the product or service – the 

social aspect of diffusion; and (2) the company marketing 

the product, through its promotional activities.  Endorse-

ment by actual users of the new product is a credible infor-

mation source for non-users.  But if the firm marketing the 

product has brand equity that creates consumer trust the 

firm’s marketing message is also credible.  Over time non-

users receive more information about the product through 

both the social system and the actions of the firm marketing 

the product.  This decreases perceived risk and increases 

the likelihood of adoption. 

Price operates on the consumer’s mind in a similar 

way.  Potential buyers are heterogeneous with respect to 

their willingness to pay for a new product because of dif-

ferences in needs, income, and the perceived product value 

(Kalish, 1985).  Because they are risk averse consumers 

discount the value of the product because of uncertainty 

over the value of the product to them specifically.  In this 

situation the seller of the product can influence diffusion.  

By pricing the product aggressively at launch, or by execut-

ing either permanent of temporary price promotions (for 

example through coupons) the seller reduces the buyers 

perceived risk and makes the buying decision easier. 

Endorsement by the social system, advertising and 

promotion, and pricing work together to reinforce the value 

of the new product offering, reduce buyer risk, and create 

new sales and increased diffusion.  As Kalish concludes: 

“The rate of adoption is therefore determined by awareness 

diffusion, which is controlled by advertising, and the rate of 

growth of the potential adopter population, which is con-

trolled by price” (Kalish, 1985). 

EXAMINING THE IMPACT OF PRICE ON MAR-

KET SIZE 

 

A wide variety of academic studies related to the im-

pact of pricing on diffusion conclude that pricing impacts 

both the rate of diffusion (both the coefficient of innovation 

and the coefficient of imitation) and the total market poten-

tial for a product category.  The impact of pricing may vary 

for durable versus non-durable goods, for products versus 

services, and for expensive versus low priced products, but 

the overall affect of pricing is to increase diffusion and 

increase market opportunity (Karine et.al., 2004). 

We earlier discussed Kalish’s finding that the combi-

nation of previous users endorsement and price directly 

addresses consumer purchase uncertainty.  Parker (1992) 

found that pricing affects the diffusion of some low priced 

durable goods.   His empirical study says that price almost 

always affects imitation, and that for some products it im-

pacts both innovation and imitation.  If price is low enough 

innovators will accept the reduced risk of being an early 

adopter, information about the product will spread more 

rapidly in the social system, creating two effects: (1) imita-

tors buy the product sooner (the shape of the curve chang-

es), and (2) more imitators are encouraged to buy the prod-

uct (the market size increases). 

Tsai et.al. (2009) studied the global LCD TV market.  

The study applied a growth model that examined the im-

pact of price on imitating behavior in diffusion of high 

technology consumer durables.  They conclude that 

“decreasing LCD TV prices stimulates the growth of inter-

nal influences and then facilitates more adoption of LCD 

TV’s.  In general the imitating tendency of potential con-

sumers is substantially driven by the price cuts”, increasing 

Figure 1B 

Standard Case of Cumulative Adopters with Market Size fixed 



 

Page 83 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

the size of the market.  The authors claim that their price 

related model “explicitly performs superior in fitness than 

the Bass Model.”   Masek (1996) observed the similar af-

fect in his study of market potential for the cable television 

industry.  His empirical findings are that price influences 

the coefficient of innovation, advertising affects the diffu-

sion rate (imitation), and expanding distribution channels 

increases the size of the market opportunity.   

Jain et.al (1990) examined “diffusion models for dura-

bles that incorporate price explicitly.”  They hypothesize 

that the price of durable goods appears to monotonically 

decline over time and that this price decline causes market 

growth.  The article applied three separate statistical mod-

els to the examination of price affects on both total market 

size (m in the Bass Model) and the rate of diffusion 

(primarily q in the Bass Model) of four different consumer 

durable products.  Below is a summary of their findings:  
 

 Room air conditioners from 1949 to 1961.  The 

authors conclude that price affected positively 

both the market size and the rate of diffusion. 

 Clothes dryers from 1949 to 1961.  Price declines 

had a “significant effect” on both the market size 

and the rate of diffusion of this product category. 

 Color televisions from 1963 to 1970.  Effects were 

similar to those observed for clothes dryers, e.g. 

significant price elasticity impacting market size 

and the rate of diffusion. 
 

Can openers from 1958 to 1970.  Here the results are 

different; price declines did not impact either market size or 

the diffusion rate.  The authors suggest that the reason is 

the low price of the product: “Price does not seem to matter 

much for this product.”  This finding suggests that there 

may be a threshold point for the affects of price on both 

market size and the rate of diffusion.  This is a topic for 

future research. 
The authors’ general conclusion from their study is 

that: 

“We see that the estimated market potential, as 

well as the eventual probability of adoption, have an 

increasing trend….  Therefore if price has a declining 

trend…the estimated market potential, as well as the 

eventual probability of adoption, increases over 

time…. (P)rice influences a consumer’s decision to buy 

the product; whereas the diffusion process determines 

the timing of the purchase.” 

 

The conclusions of Jain and Rao confirm those of 

Kmakura and Balasubramanian (1988) who also studied the 

affect of price on the market size and diffusion rate for con-

sumer durables.  They conclude that price positively affects 

both market size and diffusion: “(P)rice has a predictable 

effect on whether or not adopters are in the market”, and 

“Price does effect the diffusion of consumer durables….  

The effect of price is to increase sales from remaining 

adopters.”  Kalish (1985) reached a similar conclusion:  the 

market potential for a product category is a function of the 

price of the product and the reduction in consumer uncer-

tainty about the product because of its increased adoption.    

 

EXAMINING THE IMPACT OF PROMOTION ON 

MARKET SIZE 

 

The impact of advertising on demand and financial 

results for multiple product categories has been studied 

over multiple years by both academics and practitioners.   

However, advertising impacts on the diffusion of new 

products, on market size, and on the market share of indus-

FIGURE 2 

Impact of price on market size 



 

Page 84 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

try participants has not been fully studied (Nguyer et. al., 

2006).  Nguyer et. al. (2006) undertook a study to 

“formulate and analyze a formal model of competitive ad-

vertising strategies under market-size dynamics.”  Refer-

encing the data that Mahajan et. al. (1990) used in the pre-

viously discussed analysis of the Polaroid versus Kodak 

legal case, the authors attempted to use the Bass Model to 

determine the affect of Kodak’s and Polaroid’s combined 

advertising spending on total market demand for instant 

cameras.  Their models failed to “fit” with actual market 

demand.  They conclude that “we tend to believe that the 

rather poor performance of the… (model) may be due to 

the assumption of a static, fixed market size for this partic-

ular product.”   

Fosfur and Giarratana (2009) echo the findings of 

Nguyer et. al.  They point out that “little research has been 

conducted on the effect of rival moves along non-price 

dimensions of the market mix.”   Their empirical study of 

the affect of advertising in the carbonated soft drink indus-

try concludes that “…new advertising affects the firm’s 

financial market value through increasing total demand 

(market size dynamics)….”  Specifically: “We show that in 

the CSD market advertising increases a firm’s and its ri-

val’s financial value by increasing total demand….” 

Fosfor’s and Giarratana’s findings are consistent with 

prior studies of the affect of advertising.  Both practitioners 

and academics distinguish between generic product catego-

ry advertising and brand-specific advertising (Bass, 2005).  

These two categories of advertising have different market 

impacts.  Generic advertising increased market size by at-

tracting new customers to the product category (Berndt et. 

al., 1997).  These findings build on Lancaster’s (1984) pio-

neering study of advertising impacts, where he concluded 

that market size dynamics is the most common effect of 

advertising, suggesting that even brand advertising increas-

es demand for all market participants in multiple product 

categories.  Lancaster studied the effect of advertising on 

demand for mature consumer non-durable products, specif-

ically deodorants, shaving cream, ready-to-eat breakfast 

cereals, laundry detergents, bar soap, and cigars.   In three 

product categories (deodorants, shaving cream, and cereals) 

brand advertising increased primary demand, increasing 

sales both for the brand being advertised and for the cate-

gory in general.  For laundry detergent brand advertising 

increased total category demand with nearly all of the in-

crease accruing to the brand being advertised.  Bar soap 

and cigar advertising did not increase the total market nor 

brand market share.  This result suggests that for bar soap 

and cigars the market is not elastic, perhaps because of an 

already high household penetration for bar soap and the 

relatively limited market appeal for cigars. 

Academic studies have addressed the impact of adver-

tising on the coefficient of both innovation and imitation in 

the Bass Model as well as its affect on total market size.   

Horsky and Simon (1983) studied the diffusion of new 

banking services.  They conclude that advertising provides 

important information to innovators and suggest that the 

coefficient of innovation in the Bass Model should be a 

function of advertising with diminishing returns over time.  

Simon and Sebastian (1987) argue that advertising has 

more impact on the coefficient of imitation during the inter-

mediate stages of the product life cycle.  The effect of ad-

vertising is cumulative over the life of the product, and 

both the market size and the rate of diffusion are positively 

impacted.  This is reinforced by the previously referenced 

work of Mesak (1999) who completed an empirical study 

of the US cable industry, concluding that price influenced 

the coefficient of innovation, advertising affected the coef-

FIGURE 3 

Impact of advertising on market size  



 

Page 85 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

ficient of imitation, and additional distribution increased 

total market size.   

 

CLOSING REMARK – PRACTITIONER’S 

AND ACADEMIC PERSEPCTIVES 
 

Our extensive review of the literature in this paper, 

including academic and practitioner perspectives, provides 

compelling evidence that both price reductions and in-

creased promotional spending positively affect the total 

market opportunity for a product. This provides a strong 

case for revising the Bass Model to make the market size a 

function of the marketing mix.  In what follows, in the next 

section, we develop such a model. 

 

DEVELOPMENT OF A REVISED BASS 

MODEL INCLUDING MARKETING MIX 
 

In this section we revise the Bass Model, based on our 

findings from the review of the literature, to include the 

impact of changes in the marketing mix on market size and 

the diffusion of innovation on the demand for new prod-

ucts.  We then test the behavior of the revised model on the 

intertemporal growth in demand by running a set of simula-

tions with changes in the marketing mix with respect to 

pricing and advertising.  

To begin to develop the revised model, we start with 

the reduced form of the generalized Bass model as:  

 

St  = (pm + (q – p)Nt-1 – (q/m)Nt-1
2 )*Zt      (1) 

 

Where: 

St = number of new adopters in time period t 
p = probability of adoption by innovators (or 

“coefficient of innovation”) 
m   = total market size 
q  = probably of adoption by imitators (or 

“coefficient of imitation”). 
Nt-1  = cumulative number of previous buyers 
Zt  = time dependent term that is a function of mar-

keting mix variables 
 

The total number of adopters or total demand at time t (Nt), 

is then equal to: 

 

Nt  = St  + Nt-1          (2) 

 

Since the number of new adopters in period t (St) is a func-

tion of past levels of total demand (equation 1) the expres-

sion for the total number of adopters or the level of demand 

at time t (Nt) can be found by substituting equation 1 into 

equation 2: 

 

Nt  = pm Zt + (1+ (q – p) Zt)Nt-1 – (q/m)Zt Nt-1
2     (3) 

 

Although the generalized Bass model includes the market-

ing mix term Zt, our concern is the assumption that the 

market potential, m, is fixed and independent of the mar-

FIGURE 4 

CASE A: Impacts of a High Price Strategy with Advertising Fixed 



 

Page 86 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

keting mix. Now we propose changing the market poten-

tial, m, in the model from exogenous to endogenous and 

make it a function of the marketing mix (Zt) rather than 

including it as an independent term as shown previously in 

equations 1 and 3. 

In our revised model we use only two primary marketing 

mix variables and assume the following multiplicative ex-

pression for the market potential, where “m” is a function 

of price (P) and advertising (A): 

 

m = sP-eAf            (4) 

 

Where: 

m = market potential (size) 

s  =  scaling factor 

P =  price  

A =  advertising expenditure 

e  =  coefficient of sensitivity (elasticity) for price 

f  =  coefficient of sensitivity (elasticity) for advertising 

 

The functional form for market potential (equation 4) 

was selected because it is a standard Cobb-Douglas func-

tion which is widely used in economic studies owing to its 

robust properties with respect to stability and ease of esti-

mating parameters. The coefficient “e” represents price 

elasticity of demand; and the coefficient “f” represents ad-

vertising elasticity.  

The marketing mix, represented by only two variables 

in equation 4, can be easily extended to include any number 

of factors. We selected only two marketing mix variables in 

this paper for illustrative purposes. 

Substituting equation 4 into equation 3, which now 

includes the marketing mix term Zt as a factor influencing 

market potential, we get: 

 

Nt  = psP-eAf + (1+q – p)Nt-1 – (q/ sPeAf)Nt-1
2         (5) 

 

Given that price (P) and advertising (A) are fixed during 

the interval t, we can simplify the expression for equation 

5, the total number of adopter in the current period (Nt) as:  

 

Nt  = a + bNt-1 – cNt-1
2            (6) 

 

Where: 

a =  psP-eAf 

b =  1+q-p 

c =  q/ sP-eAf  

In this case the total number of adopters or the demand in 

the current period depends not only on the past levels of 

demand, but also on the marketing mix, i.e. the price and 

advertising levels. 

 

ILLUSTRATING THE STANDARD CASE OF THE 

BASS MODEL  

To show how changes in the marketing mix affect the 

behavior of demand in our revised Bass model, we first 

illustrate the standard case assuming the market size is 

fixed.  

To do this we utilize the Bass model parameters esti-

mated by Tigert & Farivar (1981) with respect to the adop-

tion of optical scanning equipment. Quarterly data was 

utilized and Table 1 shows the value of their estimated pa-

rameters. But the accuracy of their parameter estimates is 

not the issue in this study. The purpose here is to show how 

sensitive the demand forecasting model would be to chang-

es in the assumptions with respect to the marketing mix; as 

well as the precise nature of the change.  

 

Table 1 

Bass Parameters for Optimal Scanning  

Equipment by Tigert & Farivar (1981) 

Given these parameter estimates, the demand forecasts 

in the Bass Model behave as shown in Figure 1A. The fig-

ure shows that number of “new” adopters per quarter ini-

tially increase at an increasing rate, reach a maximum, and 

then decline. The number of new adopters per quarter even-

tually reaches zero and at this point the total (cumulative) 

number of adopters’ peak at the market size of 3290. Figure 

1B shows the cumulative or total number of adopters per 

quarter as time advances. 

 

SPECIFYING THE MARKETING MIX PARAME-

TERS 

 

Next we need to make assumptions concerning the 

parameter values of the marketing mix, equation 4. To be 

TABLE 2 

Marketing Mix Parameters in Cobb-Douglass Demand Function 

Parameters Low Medium High 

e 0.35 1.00 3.00 

f 0.25 0.50 0.75 

Bass Parameters Estimated Values 

p 0.000388 

q 0.353 

m 3,629 



 

Page 87 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

consistent with marketing theory, the value of the advertis-

ing coefficient is constrained between 0 and 1. This con-

straint reflects the common finding that there are diminish-

ing returns to advertising.  With respect to price elasticity, 

Feenstra (2003) notes that there has been a “blizzard” of 

studies over the past decade on this topic.  In a well de-

tailed study, Foekens, et. Al (1998) estimates the price elas-

ticity to range from inelastic (-0.33) to highly elastic (-

3.15), which is consistent with the findings of others.  As a 

result, we test the impacts of low, medium and high values 

for both price elasticity and advertising values as specified 

in Table 2.   

The scaling factor, s, was set to yield a pre-specified 

market size of 3290, at the middle range for price and ad-

vertising levels, to be consistent with the Tigert & Farivar 

(1981) study used to select the parameters of the Bass mod-

el (specified in Table 1). Given these parameter estimates, 

the relationship between price and advertising on market 

size is illustrated in Figures 2 and 3. In both of these Fig-

ures the price and advertising levels are normalized for ease 

of illustration and comparison, where an index value of 100 

represents the median value.  

In Figure 2 the high price elasticity of demand is a 

relatively flatter curve, showing that the impact of price on 

the total market size increases much more rapidly as price 

declines relative to the low price elasticity. (Although slope 

is not the same as price elasticity, for any given price range, 

the flatter the slope the greater the elasticity.) For example 

at a price index of 100 the market size is 3200 for both 

curves. If price drops to 80, the market size of the high 

elasticity of demand doubles to about 6400, while the low 

price elasticity increases by only 400 units to 3600. 

In Figure 3 the high advertising elasticity of demand is 

a relatively flatter curve, showing that the impact of adver-

tising on the total market size increases much more rapidly 

with increases in advertising relative to the low advertising 

elasticity. Also, there are diminishing returns to advertis-

ing. As advertising increases we see that the low elasticity 

demand curve becomes much steeper (more vertical) indi-

cating that the gain in market demand eventually becomes 

very small as advertising increases. For example at an ad-

vertising index of 100 the market size is about 3200 for 

both curves. If advertising increases to 200, the market size 

of the high elasticity of demand almost doubles, increasing 

about 2800 units to a market size of 6000, while the low 

price elasticity increases by only 800 units to 4000. 

 

TESTING THE IMPACT OF THE MAR-

KETING MIX ON REVISED BASS MODEL 
 

In this study twenty-eight computer simulations were 

run of the revised Bass Model with changes in the pre-

specified levels of both price and advertising, given varying 

assumptions with respect to the elasticity coefficients in the 

marketing mix function. In particular, sensitivity analysis 

was done on the following scenarios: 

The impact of a high price strategy given relatively 

high and low price elasticity coefficients with advertising 

fixed. 
The impact of a low price strategy given relatively 

high and low advertising elasticity coefficients with adver-

FIGURE 5 

CASE C: Impact of High Advertising with Price fixed 



 

Page 88 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

tising fixed 
The impact of a high advertising strategy given rela-

tively high and low advertising elasticity coefficients with 

price held constant. 
The impact of a low advertising strategy given rela-

tively high and low advertising elasticity coefficients with 

price held constant. 
The impact of a 2% quarterly growth rate in advertis-

ing with high and low advertising elasticity. 
The impact of a 2% quarterly growth rate in both ad-

vertising and price with high and low advertising elasticity. 
To be succinct, only cases A, C, E and F are illustrated 

in Figures 4 to 7 below but all cases are discussed. Case B 

is discussed along with the findings of case A; and case E 

is discussed with the findings of case F. 

Case A shows that a high price will dampen the speed 

of diffusion, and that the higher the price elasticity the 

greater the dampening effect. Adopters per quarter reach a 

peak of almost 300 in the standard case, but only 265 with 

low elasticity, and only 150 with high elasticity. Case B 

(not illustrated in a figure) provides the same results but in 

the opposite direction. A low price will heighten the speed 

of diffusion, and it will be greatest with a high elasticity. 

Case C shows that a high advertising level will in-

crease the speed of diffusion; and the higher the advertising 

elasticity the greater the effect. Adopters per quarter reach 

a peak of only 290 in the standard case, but the peak in-

creases to 320 with high advertising even if advertising 

elasticity is low; and increases even further to 350 with 

high elasticity of advertising. Case D (not illustrated in a 

figure) provides consistent results but in the opposite direc-

tion. A low advertising level will dampen the speed of dif-

fusion, and the higher the advertising elasticity the greater 

dampening effect.  

Case E shows the impact of advertising growing at a 

rate of 2% per quarter. The significant finding in this case 

is that the length of time of diffusion increases before the 

peak is reached along with increases in the speed of diffu-

sion and the peak level of demand (quarterly adopters).  

This finding is independent of the level of the elasticity of 

advertising. In the standard case, the peak is reached after 

23 quarters at a level of 300 adopters. But with growth in 

advertising, coupled with a high elasticity of advertising, 

the peak is not reached until 25 quarters at a peak demand 

level of 500 adopters per quarter.  

The final set of simulations, case F, shows the impacts 

of growth in both advertising and price of 2% per quarter. 

Here the rate of diffusion and the peak level of demand that 

is reached are dependent on the level of the advertising 

elasticity. If the advertising elasticity is high enough the 

rate of diffusion will increase over the standard case. But 

an interesting finding is that the rate of decline is higher 

after the peak is reached.  

 

CONCLUSION 
 

There is growing interest in the modeling and forecast-

ing of demand for new and innovative products owing to 

the rapid pace of technological change. For this purpose, 

the Bass model is a widely respected and utilized diffusion 

FIGURE 6 

CASE E: Impact of growth in Advertising with Price fixed 



 

Page 89 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

model. But one key element of the model has been brought 

into questions in our research, that is, the assumption that 

the market potential is constant, even with changes in the 

firm’s marketing mix. There is compelling evidence from 

both academic studies and anecdotal evidence from practi-

tioners that price reductions and increased promotional 

spending positively affect the total market opportunity for a 

product.  It is recommended that those who develop models 

to forecast demand and those involved in the design and 

development of business simulations should consider the 

impact of the market mix in general, and pricing and pro-

motion specifically, on both the total market opportunity 

for a product and the shape of the diffusion curve.   

In this paper we have revised the Generalized Bass 

model to include the complexities of modeling the effects 

of the marketing mix on market size. The simulations de-

veloped from our revised model show that high prices will 

slow down both the rate of diffusion and level of demand; 

while high advertising levels will increase the speed of 

diffusion and level of demand. The precise nature of the 

impacts is influenced not only by the level of price and 

advertising but also on the elasticity of the marketing mix 

variables. High price elasticity will increase the speed of 

diffusion when prices are kept low but will dampen the 

speed of diffusion when price is high. In contrast, high ad-

vertising elasticity will quicken the speed of diffusion inde-

pendent of the growth in advertising.  

These conclusions have important implications for the 

strategies of firms competing in specific product markets.  

Using price reduction and/or increased promotional spend-

ing as tactics to increase market size is a rational marketing 

strategy if one or both of the following criteria is achieved:  

 The market size is sufficiently elastic so that the in-

creased demand from the price reduction overcomes 

the financial exposure associated with the lost revenue 

and margin from customers who would have pur-

chased the product at the higher price.   

 The market size is elastic in a similar way to increased 

promotional spending.  Promotional investments in-

creased the awareness of potential buyers for a prod-

uct, and reduce buyer’s risk by providing information 

about the product.  Increase promotional spending is 

justified when each additional $1 spent on promotion 

provides a satisfactory return to the firm. 

Importantly both of these criteria may require that the 

firm make additional investments in capital (to increase 

production capacity), in operating expenses (to acquire 

additional components from suppliers and to invest in the 

marketing mix), and in people (to manage both the in-

creased demand and the cost-related process improve-

ments).   

Mahajan et. al. (1990) observed that “The diffusion 

process consists of four key elements: innovation, commu-

nication channels, time, and the social system.”  This con-

clusion is only partially correct.   Our research suggests that 

management of the marketing mix may impact all the key 

elements of the Bass Model: the coefficients of innovation 

and imitation and, important, the total market size.  The 

focus of our study is the impact of price and promotion on 

the market opportunity for a product category as represent-

ed in the Bass Model.  We conclude that price and promo-

tion generally have a positive impact on market size.  There 

FIGURE 7 

CASE F: Growth in both Advertising and Price 



 

Page 90 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

are two caveats related to this conclusion: 

 Changes in pricing and advertising may not have any 

effect on the market size for certain product categories.  

We hypothesize that where the consumer already per-

ceives that the product price is low (as in the case of 

electric can openers) or when the market itself in ine-

lastic because of the nature of the product (as in the 

case of cigars) the total market opportunity may not 

increase as a result of pricing or advertising actions by 

market participants. 

 The degree of the impact of price and promotion on the 

increase in market size likely is not uniform and varies 

by product category.  For example pricing action or 

advertising campaigns for desirable consumer products 

such as electronics (iPad as an example) or fashion 

(UGGS as an example) may result in substantial in-

creases in both market size and diffusion, while similar 

actions for consumer or commercial durables may 

have a more modest effect.   

These findings are significant with respect to the effec-

tive design of business simulations. If a business simulation 

were designed to accept market size as a constant, as in the 

Bass model, there would be a zero sum game. In this case, 

the only way a firm would be able increase demand is by 

taking market share from competitors. If correct a com-

pletely static market would result; and in hyper-competitive 

markets, lowering prices and margins of market partici-

pants.  In this scenario, firms would have less incentive to 

invest in innovations or increased capacity, potentially 

causing market stagnation in the product category.  The 

idea or assumption of a fixed market size would certainly 

send the wrong signal to business strategists, and to stu-

dents learning experientially from business simulations. 

 

REFERENCES 
 

Bass, F. M. (1969). "A New Product Growth Model for 

Consumer Durables," Management Science, 15 

(January), pp 215-227. 

Bass, Frank M, and Gordon, Kent, Ferguson, Teresa, 

Githens, Mary Lou (2001). “Direct TV: Forecasting 

Diffusion of a New Technology Prior to Product 

Launch”, Interfaces 31:3, Part 2 (May-June), pp S82-

S93.   

Bass, F.M., and A Krishnamoorthy, A. Prasad, S.P. Sethi 

(2005). “Generic and Brand Adversiting Strategies in a 

Dynamic Duopoly”, Marketing Science, 24 (4), 556-

568. 

Berndt, E.R., and T. M. Bui, D.H.L. Lucking-Reiley, G. L. 

Urban (1997). “The roles of marketing, product quali-

ty, and price competition in the growth and composi-

tion of the US anti-ulcer drub industry, The Economics 

of New Goods, Unversity of Chicago Press, Chicago, 

277-322. 

Cannon, Hugh M, Laura C. McGowan, and Sung-

JoonYoon, (1994).”Incorporating Advertising Strategy 

into Computer-based Business Simulations”,  Develop-

ments in Business Simulations and Experiential Learn-

ing, 21, 57-62. 

Cannon, Hugh M., John D. Leckenby, and Avery M. Aber-

nethy (1996), “Modeling Advertising Effectiveness”, 

Developments in Business Simulations and Experien-

tial Learning, 23, 1-7. 

Carvalho, Gerald(1991). “Theoretical Derivation of a De-

mand Function for Business Simulators, De-

velopments in Business Simulations and Experiential 

Learning, 18, 11-14. 

Chandrasekaran, Deepa and Bellis, Gerard (2007) “A Criti-

cal Review of Marketing Research on Diffusion of 

New Products”, Review of Market Research, Vol. 3, 

Emerald Group Publishing Limited , pp 39-80. 

Feenstra, Robert C. and Shapiro, D. (2003). Scanner Data 

and Price Indexes, University of Chicago Press, ISBN 

0-226-23965-9 (Volume URL: http://www.nber.org/

books/feen03-1) 

Foekens, Eijte W, Leeflang, Peter S.H., and Wittink, Dick 

R. (1998). “Varying parameter models to accommo-

date dynamic promotion effects”, Journal of Econo-

metrics, 89, 249-268. (link to article is at: http://

www.google.com/search?

q=price+elasticity+of+optimal+scanning+equipment&

ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-

US:official&client=firefox-a) 

Fosfur, Andrea and Giarratana, Marco S (2009). “Master of 

War: Rivals’ Product Innvoation and New Advertising 

in Mature Product Markets”, Management Science, 55 

(February), No. 2.  pp 181-191. 

Gold, Steven C. and Thomas F. Pray (1983). “Simulating 

Market and Firm Level Demand: A Robust Demand 

System”, Developments in Business Simulations and 

Experiential Learning  NEED COMPLETE CITE 

HERE 

Gold, Steven C. and Thomas F. Pray (1998), 

“Technological change and intertemporal movements 

in consumer preferences in design of computerized 

business simulations with market segmentation,” De-

velopments in Business Simulation and Experiential 

Learning, Vol. 25: 156 - 167. 

Goosen, Kenneth R. (1986), “An Interpolation Approach to 

Developing Mathematical Functions for Business Sim-

ulations,” Developments in Business Simulation and 

Experiential Exercises, 13, 248-255. 

Goosen, Kenneth R. (2010), “The Effect of Advertising on 

Demand in Business,” Developments in Business Sim-

ulation and Experiential Exercises, 37, 79-94. 

Goosen, Kenneth R. (2011), “The Meaning of Firm De-

mand in Business Simulations,” Developments in 

Business Simulation and Experiential Exercises, 38, 

375-381. 

Gourville, John T. (2006) “Eager Sellers and Stony Buy-

ers”, Harvard Business Review, 84(6), June, PP 98-

106. . 

Horsky, Dan and Simon, Leonard S. (1990). “Advertising 

and the Diffusion of New Products,” Management 

http://www.nber.org/books/feen03-1
http://www.nber.org/books/feen03-1
http://www.google.com/search?q=price+elasticity+of+optimal+scanning+equipment&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
http://www.google.com/search?q=price+elasticity+of+optimal+scanning+equipment&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
http://www.google.com/search?q=price+elasticity+of+optimal+scanning+equipment&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
http://www.google.com/search?q=price+elasticity+of+optimal+scanning+equipment&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
http://www.google.com/search?q=price+elasticity+of+optimal+scanning+equipment&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a


 

Page 91 - Developments in Business Simulation and Experiential Learning, volume 39, 2012 

Science, 1 (Winter), 1-18. 

Jain, Dipak C. and Rao, Ram C. (1990). “Effect of Price on 

the Demand for Durables: Modeling, Estimation, and 

Findings”, Journal of Business and Economic Statis-

tics, Vol 8 (April), No. 2, pp 163-170. 

Jamieson, Linda and Frank Bass “Adjusting Stated Inten-

tion Measures to Predict Trial Purchase of New Prod-

ucts: A Comparison of Models and Methods”, Journal 

of Marketing Research, Vol. 26 

Jamieson, Linda and Tigert, Douglas, Farivan, Behrooz 

(1981) “The Bass Model product growth model: A 

sensitivity analysis for a high technology product”, 

Journal of Marketing, Vol. 45 (Fall), pp 81-90 

Kalish, Shlomo (1985). “A New Product Adoption Model 

With Pricing, Advertising, and Uncertainty,” Manage-

ment Science, 31 (December) 1569-85. 

Kamakura, W. and Balasubramanian, S. (1988). “Long 

Term View of the Diffusion of Durables: A Study of 

the Role of Price and Adoption Influences Processes 

Via Test of Nested Models”, International Journal of 

Research in Marketing, 5, pp 1-13. 

Kim, W. Chan and Renee Mauborgne (2005). “Blue Ocean 

Strategy: From Theory to Practice:, California Man-

agement Review, Vol. 47 No. 3 (Spring), pp 105-121. 

Kotler, Philip, and Keller, Kein Lane (2006). “Marketing 

Management”, 12th edition.  Pearson Prentice Hall, pp 

567-591. 

Lancaster, K.M. (1984). “Brand advertising competition 

and industry demand”, Journal of Advertising, 13 (4), 

pp 19-30. 

Mahajan, Sharma, and Buzzell (1993). “Assessing the Im-

pact of Competitive Entry on Market Expansion and 

Incumbent Sales,” Journal of Marketing, Vol. 57 

(July). Pp 39-52. 

Mahajan, Vijay, Muller, Eitan, Bass, Frank (1990). “New 

Product Diffusion Models in Marketing: a Review and 

Directions for Research.”  Journal of Marketing, Vol. 

54 (January), 1-28. 

Mahajan, Vijay, Sharma, Subhash, & Buzzell, Robert D, 

“Assessing the Impact of Competitive Entry on Market 

Expansion and Incumbent Sales”, Journal of Market-

ing, Vol. 57 (July), pp 39-52. 

Mesak, Han I (1999). “Incorporating price, advertising, and 

distribution in diffusion models of innovation: Some 

theoretically and empirical results,” Computer and 

Operations Research, Volume 23, Issue 10, October, 

pp 1007-1023. 

Nguyen, Dung and Shi, Lei, (2006) “Competitive Advertis-

ing Strategies and Market-Size Dynamics: a Research 

Note on Theory and Evidence,” Management Science, 

Vol 52 No. 6, (June), pp 965-973. 

Ofek,  Elie (2005) “Forecasting the Adoption of a New 

Product, Harvard Business School Press # 9-505-062. 

Robinson, Bruce and Lakhani, Chet (1975). “Dynamic 

Price Models for New Product Planning,” Manage-

ment Science, 10 (June), 1113-22. 

Rogers, Everett M (1983). Diffusion of Innovations, 3rd 

edition, New York: The Free Press.   

Simon, Herman and Sebastian, Karl-Heinz (1987). 

“Diffusion and Advertising: The German Telephone 

Company,” Management Science, 33 (April), 451-66. 

Thavikulwat, Precha (1988), “Simulating Demand in an 

Independent-Across-Firms Management Game,” De-

velopments in Business Simulation and Experiential 

Exercises, 15, 183-187. 

Teach, Richard D., 1990, “Demand Equations Which In-

clude Product Attributes, Developments in Business 

Simulations and Experiential Learning, Vol. 17   

NEED PAGE NUMBERS 

Trichy V Krishnan; Frank M Bass; V Kumar , Impact of a 

late entrant on the diffusion of a new product/service,  

Journal of Marketing Research; May 2000; 37, 2; 269-

276. 

Tigert & Farivar (1981). The Bass New Product Growth 

Model: A sensitivity analysis for a high Technology 

product, Journal of Marketing, Vol. 45, 81-90. 


	Table of Contents
	Volume 39, 2012
	Designing the Training Challenge
	Follow The Leader: Are we Teaching our Students to be Thinkers or Followers?
	Two Free-Rider-Accepting Methods of Organizing Groups for a Business Game
	Additional Benefit Through Competency Models
	Assessing Brand Portfolio Normative Consistency & Trends With The Normative Position of Brands & Trends Package
	Modeling the Impact of Marketing Mix on the Diffusion of Innovation in the Generalized Bass Model of Firm Demand
	Play it Forward! The Design and Development of a Forward Contract Simulation
	Positioning the Company: Increasing Profits in Social Networks
	Merger of Companies in Business Game Exercise
	Towards a Knowledge-Based Approach for Autonomouse Trading Agent
	An Exploratory Study of the Impact of a Simulation Exercise on the Managerial and Personality Traits and the Decision Making Styles of Marketing Students
	Should the Concept of Potential Customers be the Foundation of Demand Theory in Business Simulations?
	Teaching Sustainability Experientially
	Drawing Upon Experience and Research to Improve Future Communications
	Improving Assessments of Student Learning Outcomes (SLO) Over Time
	The Effect of Affective Domain Characteristics on Behavioral or Psychomotor Outcomes
	Gossip?  No, Not Me!  An Experiential Exercise
	Student Advisement Using Gantt Charts: An Experiential Exercise in Management Theory
	Practicing Teachers as Digital Game Creators: A Study of the Design Considerations
	Designing and Solving Crossword Puzzles: Examining Efficacy in a Classroom Exercise
	Difficult Times Call for Innovative Measures: Microfinance as Experiential Learning in Higher Education
	Catalysts, Client Services, and Community Change: Interdisciplinary Collaboration in a Nascent Microfinance Initiative
	Build A Business . . . In An Hour or Less: Getting Closer to Reality into the Classroom
	Smart Goals: How the Application of SMART Goals can contribute to achievement of Student Learning Outcomes
	The Use of Data in "Live" Cases to Encourage Systems Thinking and Integrative Analysis: An Exercise Linking Human Resource Programs and Financial Outcomes in Real Organizations
	Experiential Education as a Process of Changing Mental Frames by Inducing Insight Learning
	Process and Content Integration in an Experiential Learning Guided Internship Program
	Good-bye Discussion Thread:  Creating a Community of Inquiry in an Online Master's Program
	Fiction as a Constructivist Tool for Learning Process Consultation in an Online Environment: Shaping the Context, Introducing the Dialogue
	Can Simulations Provide a Better Experience? A Capstone Application
	Modeling a Modest Proposal for Increasing the Efficiency of Academic Reserarch Dissemination
	Experience GEO: A Massively Multiplayer Game
	SysTeamsGames Three Games for Management Simulation
	SimVenture - A Start-Up Business Simulation
	Stellarbucks Simulation Developing Games Using Strategy Maps and Balanced Scorecards
	Strategy Dynamics Models - Powerful But Simple In-Class Games
	Simulating Scenarios for Financial Statement Analysis
	A Valuation Model of the Simulated Firm
	Writing the Land: An Interdisciplinary Experiential Approach
	On the Estimation of the Probability of Meeting Financial Commitments: A Behavioraial Finance Perspective Using Business Simulations