St. Petersburg University
Graduate School of Management
Master in Information Technologies and Innovation Management
DETERMINANTS OF INNOVATION
PERFORMANCE IN RUSSIAN COMPANIES
Master’s Thesis by the 2nd year student
Concentration — Master in Information
Technologies and Innovation Management
Artemii Uverskii
Research advisor:
Associate Professor, Irina V. Berezinets
St. Petersburg
2016
1
ЗАЯВЛЕНИЕ О САМОСТОЯТЕЛЬНОМ ХАРАКТЕРЕ ВЫПОЛНЕНИЯ
ВЫПУСКНОЙ КВАЛИФИКАЦИОННОЙ РАБОТЫ
Я, Уверский Артемий Алексеевич, студент второго курса магистратуры направления
«Менеджмент», заявляю, что в моей магистерской диссертации на тему «Детерминанты
инновационной продуктивности в Российских компаниях», представленной в службу
обеспечения программ магистратуры для последующей передачи в государственную
аттестационную комиссию для публичной защиты, не содержится элементов плагиата.
Все прямые заимствования из печатных и электронных источников, а также из
защищенных ранее выпускных квалификационных работ, кандидатских и докторских
диссертаций имеют соответствующие ссылки.
Мне известно содержание п. 9.7.1 Правил обучения по основным образовательным
программам высшего и среднего профессионального образования в СПбГУ о том, что «ВКР
выполняется индивидуально каждым студентом под руководством назначенного ему
научного руководителя», и п. 51 Устава федерального государственного бюджетного
образовательного
учреждения
высшего
профессионального
образования
«Санкт-
Петербургский государственный университет» о том, что «студент подлежит отчислению из
Санкт-Петербургского
университета
за
представление
курсовой
или
выпускной
квалификационной работы, выполненной другим лицом (лицами)».
_______________________________________________ (Подпись студента)
________________________________________________ (Дата)
2
STATEMENT ABOUT THE INDEPENDENT CHARACTER
OF THE MASTER THESIS
I, Artemii Uverskii, second year master student, program «Management», state that my
master thesis on the topic «Determinants of innovation performance in Russian companies», which
is presented to the Master Office to be submitted to the Official Defense Committee for the public
defense, does not contain any elements of plagiarism.
All direct borrowings from printed and electronic sources, as well as from master theses,
PhD and doctorate theses which were defended earlier, have appropriate references.
I am aware that according to paragraph 9.7.1. of Guidelines for instruction in major
curriculum programs of higher and secondary professional education at St.Petersburg University «A
master thesis must be completed by each of the degree candidates individually under the supervision
of his or her advisor», and according to paragraph 51 of Charter of the Federal State Institution of
Higher Professional Education Saint-Petersburg State University «a student can be expelled from St.
Petersburg University for submitting of the course or graduation qualification work developed by
other person (persons)».
________________________________________________(Student’s signature)
________________________________________________ (Date)
3
АННОТАЦИЯ
Автор
Уверский Артемий Алексеевич
Название магистерской «Детерминанты инновационной продуктивности в Российских
диссертации
компаниях»
Факультет
Высшая школа менеджмента
Направление
080200
подготовки
технологии и инновационный менеджмент)
Год
2016
Научный руководитель
Ирина Владимировна Березинец, к.ф.-м.н., доцент
“Менеджмент”
(Профиль:
Информационные
Описание цели, задач и Цель исследования состоит в определении детерминант
основных результатов
инновационной продуктивности в Российских компаниях. Для
достижения поставленной цели, на основе теоретического
бекграунда
была
взаимосвязи
между
разработана
теоретическая
инновационной
модель
продуктивностью
компании и рядом детерминант. Тестирование теоретической
модели
было
проведено
на
основе
многофакторной
регрессионной модели. Эмпирическое исследование было
основано на выборке из 148 Российских компаний, имеющих
разный размер и отраслевую принадлежность. Регрессионный
анализ показал, что имеет место положительная взаимосвязь
между затратами на НИОКР, маркетинг, характеристиками
инновационной стратегии компании и продуктовой инновацией
в компании, а также отрицательная взаимосвязь между
затратами на обучение сотрудников и продуктовой инновацией
в
компании.
взаимосвязь
Кроме
между
того,
имеет
место
характеристиками
положительная
инновационной
стратегии компании, затратами на обучение сотрудников и
процессовой инновацией в компании.
Ключевые слова
Инновации,
модель
инноваций
на
уровне
фирмы,
детерминанты инновационной продуктивности
4
ABSTRACT
Master Student's Name
Artemii Uverskii
Master Thesis Title
“Determinants of innovation performance in Russian companies”
Faculty
Graduate school of management
Main field of study
080200 “Management” (specialization: Information Technologies
and Innovation Management)
Year
Academic
2016
Advisor’s Irina V. Berezinets, PhD in Physico-mathematical sciences,
Name
Associate Professor
Description of the goal, The goal of the research is to identify determinants of innovation
task and main results
performance in Russian companies. To achieve this goal, the
theoretical model was constructed on the basis of thorough review of
existing studies. The model outlined relationship between the
company’s innovation performance and a set of determinants.
Theoretical model was tested using multiple linear regression
analysis on the sample of 148 Russian companies of different size
and industry. The regression analysis identified positive correlation
between the company’s R&D expenses, marketing expenses, features
of innovation strategy and product innovation, as well as negative
correlation between expenditures on employees training and product
innovation. Furthermore, there is a positive correlation between
expenditures on employees training, features of innovative strategy
and process innovation.
Keywords
Innovation, model of innovation at the firm’s level, determinants of
innovation performance
5
Table of contents
Introduction........................................................................................................................ 7
1
2
Theoretical research on determinants of innovation performance ........................ 9
1.1
Nature, definition and taxonomy of innovation .......................................................... 9
1.2
Diffusion of innovation ............................................................................................. 14
1.3
Innovations in companies ......................................................................................... 24
1.4
Justification of AMI model determinants ................................................................. 36
1.5
Methodology ............................................................................................................. 49
Results of the regression analysis, discussion and conclusion ............................... 59
2.1
Statistical results of the regression analysis .............................................................. 59
2.2
Discussion of the results ........................................................................................... 60
2.3
Theoretical contribution, limitations and further research ........................................ 66
2.4
Practical contribution ................................................................................................ 68
3
References .................................................................................................................. 69
4
Appendices ................................................................................................................. 74
6
INTRODUCTION
Innovation is important for companies and economies. It makes the company more
competitive, creates new markets and helps to reduce costs. Innovation is a source of competitive
success (Drucker 1985). Crepon, Duguetb and Mairessec (2006) created and tested a model that
proved that firm productivity is positively correlated with the company’s innovation performance.
Consultancy companies, for instance, PwC recognize innovation as a driver for rapid revenue grow
and a factor to maintain long-term enterprise growth (PwC 2013). The economy of the country also
benefits from innovations. Many years ago, Schumpeter (1942) explained that innovation is a force
that help the economy to progress: innovation incessantly revolutionizes the economic structure
from within, incessantly destroying the old one, incessantly creating a new one. Even though Russia
was not so rapid in understanding of importance of innovation, our country nowadays pays more
and more attention to it. For instance, the Russian government will soon introduce the law that will
oblige state-owned companies to implement innovations and order research projects in Russian
educational and scientific centers. Moreover, the success of innovation programs in the state-owned
companies, such as Aeroflot, RZD, and Gazprom, will affect at least 10% of total executives’ bonus
(Kommersant 2016).
This rearches aims to understand what factors are correlated with the company’s innovation
performance. Since the 1950s, there has been a proliferation of innovation models, each purporting
to explain and/or guide the process of innovation within industrial firms. Rothwell (1994) analysed
of state-of-the-art models of innovation processes at the firm level and classified these models into
five generations in his article “Towards the Fifth‐generation Innovation Process”. He identified
several points of weaknesses in these models. First, previous models provide one best way of
innovation process, eliminating alternative paths that do exist. Second, most of the models assume
that the companies behave too rationally, being able to hypothesize a solution to an innovation
problem, such as a new product development, and then systematically solve the problem, using a
standard toolkit such as design thinking, prototype testing and market research – the assumption that
does not always met. Third, the models lack a coherent theoretical base, which is important because
it can help to put innovation within the wider organizational and strategic context in which it
belongs. Forth, all the models deal with innovation leaders, neglecting latecomers. This fact is
especially important for scientist who research innovation in developing countries, e.g. Russia, since
many companies there do not develop innovations themselves, but rather adopt them from abroad.
Fifth, the majority of models deals describe processes in the large corporations, not paying attention
7
to medium and small companies, where innovation process usually do not have any formal stages
and domain.
In order to address the issues with previous models, the attributive model of innovation
(AMI) was created. This model outlines what qualities the company should have to be successful in
innovation. The main difference with the previous models is that the attributive model does not aim
to depict any particular process that companies should follow, but rather identify key attributes that
the company should have to be successful in innovation. For instance, a small company may not
even have R&D department. Therefore, for such companies it is definitely inappropriate to apply
existing models. The attributive model fits both innovation leaders and latecomers, and both large
corporations and small companies, because the determinants of AMI neglect organizational structure
and encompass sources of innovations far beyond only R&D department.
In order to prove the attributive model statistically, a linear regression was conducted in IBM
SPSS on cross-sectional data about 148 Russian companies. The regression identified correlation
between some determinants of innovation performance (company’s spending on R&D, training, and
marketing preparation for innovative products; efficiency of the company’s innovation strategy) and
the company’s innovation performance (product innovation, process innovation). However,
additional studies are required to establish causality besides correlation.
8
1 THEORETICAL
RESEARCH
ON
DETERMINANTS
OF
INNOVATION PERFORMANCE
1.1 Nature, definition and taxonomy of innovation
In order to study innovation successfully, we need to come up with common definition of
this phenomenon first, since there is no common definition of innovation neither abroad, nor in
Russia. For instance, Maryanenko (2008) provided the results of the survey conducted among
managers: the definition of innovation varied from manager to manager: 26% of respondents
considered innovation a solution that identifies and address unsatisfied customer needs, 23% - a
progress or advancement that enables to do something better, and 5% - a discovery or something
connected with scientific revolution. Not only practitioners have in mind different definitions of
innovation, but also scientists. To solve this problem, some scientists conducted research on
innovation definition. For example, Cumming (1998) conducted a survey in which he analyzed
definitions of innovation that were published from the end of 1960s. He came up with two broad
groups of definitions, each of which used the following basis: first, something new - invention, idea
or concept; second, something sellable - implementation or commercialization of the added value.
Since there are a great number of opinions what exactly innovation is, but we need to be precise in
this discussion, for the theoretical discussion the definition of innovation as “something new” was
chosen.
Rycroft and Kash (2000) outlined three types of innovations based on their strength and
influence: incremental, major and fundamental. These three types and their influence on industry
performance dynamic is depicted on Figure 1.
9
Figure 1. Three types of innovations. Source: Rycroft and Kash (2000)
First, incremental innovations. These are gradual development of existing technologies and
goods. These innovations are not an outcome of a purposeful full-scale research, but rather some
enhancement of existing technologies. Despite these innovations increase productivity that leads to
competitive advantage, they do not change production and marketing dramatically.
Sandeep Kishore uses iPhone as an example of incremental innovation in his article “The
power of incremental innovation”, published in Wired magazine:
One of the most successful and recent examples of incremental innovation is the
iPhone. While smartphones existed before Apple entered the market, it was mostly
the incremental innovations of a larger touchscreen, the app store, various ease of use
and an improved overall experience, which enabled the iPhone to be the first in
making smartphones mainstream.
Apple then created a whole new ecosystem which made the iPhone a preferred
medium for accessing the internet, sending e-mail, finding directions, playing games,
conducting online transactions and generally becoming a central part of our daily
10
lives. Last year, it shipped 125 million iPhones. Incremental innovation has brought a
fundamental change in our behavior and created a market that will be worth $1.6
trillion by 2018.
Uber is another example of incremental innovation. Uber is a taxi company that enables
customers to order a taxi through the mobile app. It is much cheaper and convenient than traditional
taxi service: the waiting time usually is less than eight minutes, whereas it is around twenty minutes
for traditional taxi; the fare is around half as much as that of the traditional taxi. Moreover, you can
use Uber even if you do not have neither cash nor a bankcard with you: Uber charges your bank
account without physical contact with your card like off-line store, but rather like on-line store,
using the card’s data.
Incremental innovation contributes much into the global economy development. Bain and
Company (2011) estimates contribution of incremental innovations to global GDP by 2020 as $5
trillion. Moreover, Bain regards this incremental innovation, or “soft innovations” in Bain’s terms,
as one of the eight macro trends that will propel global economic growth over until 2020 by
“changing our basic habits, from the way we drink coffee (think mochaccinos rather than drip brew)
to the way we buy clothes (with matching outfits delivered to our doorstep rather than shopped for
piecemeal in stores)”.
Second, major innovations. These are substantial for science evolution inventions or
discoveries, but they are one-offs, not systematic ones. Despite they can considerably change market
power of existing technologies or create new market leaders, radical innovations cannot significantly
affect the entire economy of even any particular industry to elicit technology paradigm shift. E-mail,
search engines such as Google and Yandex, social networks, and Wi-Fi illustrate the concept of
major innovation. These technologies redefine the competitive landscape, but not as much as to
reshape the whole economy or industry.
Third, fundamental innovations. These innovations lead to technology system changes and
elicit drastic changes of existing technologies. These changes influence entire technology clusters,
crowding out whole product categories and their producers, and provide particular companies with
sustainable competitive advantage through excessive added value.
11
Heather Whipps diligently describes the impacts of steam engine, which is an example of
fundamental innovation, on the world’s economy and history in his article “How the Steam Engine
Changed the World”, published in Live Science on June 16, 2008:
The simultaneous perfection of the steam engine and the beginning of the Industrial
Revolution is a chicken and egg scenario that historians have long debated. The
world was becoming an industrialized place before the advent of steam power, but
would never have progressed so quickly without it, they argue.
Factories that still relied on wind or waterpower to drive their machines during the
Industrial Revolution were confined to certain locales; steam meant that factories
could be built anywhere, not just along fast-flowing rivers.
Those factories benefited from one of the world's greatest partnerships — that of
Watt and Matthew Boulton, a British manufacturer. Together, they tailored Watt's
steam engine to any company that could use it, amassing great fortunes for
themselves but also sharing research over vast distances.
Transportation was one of those important beneficiaries. By the early 1800s, highpressure steam engines had become compact enough to move beyond the factory,
prompting the first steam-powered locomotive to hit the rails in Britain in 1804. For
the first time in history, goods were transported over land by something other than
the muscle of man or animal.
Another example of fundamental innovation is the invention of an airplane. An airplane
significantly reduced travel time, enabling more rapid logistics of goods and people. This increase in
speed affected not only international trade and economy, but also the perception of speed of life by a
mankind, the fact frequently mentioned in belles-lettres.
Major and fundamental innovation concepts are highly connected with the idea of creative
destruction, proposed by Schumpeter (1942) in his book “Capitalism, Socialism and Democracy” to
explain how the economics is developed under capitalizm. Creative destruction is a “process of
industrial mutation that incessantly revolutionizes the economic structure from within, incessantly
destroying the old one, incessantly creating a new one”. Schumpeter provides the example of his
concept, using retail industry, in which “the competition that matters arises not from additional
shops of the same type, but from the department store, the chain store, the mail-order house and the
12
supermarket which are bound to destroy [traditional stores] sooner or later”. Here something new
(supermarkets) destroys something old (traditional stores), whereby revolutionizing the economic
structure from within. Another example to illustrate the concept of creative destruction is invention
of personal computers: introduction of personal computers, led by Intel and Microsoft, destroyed
many mainframe computers companies, but revolutionized the whole industry and advanced the
whole economy.
Definition and classification of innovations by Oslo Manual, which was used for the
research, were created to be used in standardized surveys of firms. According to Oslo Manual, “an
innovation is the implementation of a new or significantly improved product (good or service), or
process, a new marketing method, or a new organizational method in business practices, workplace
organization or external relations (OECD 2005)”. It does not matter whether the company created
something new in-house, or adopted it from outside (e.g. through patent acquisition). What matters
is that the company should implement this “something new” to consider it innovation. The manual
distinguishes four different types of innovation: product, process, marketing and organizational
innovations. These types and their definitions are outlined on the Figure 2. Since this research was
restricted in time and resources, only product and process innovation in Russian companies were
studied.
Figure 2. Four types of innovation. Source (OECD 2005)
13
1.2 Diffusion of innovation
Rogers (1962) proposed a theory that explains the diffusion of innovation through cultures.
In his book “Diffusion of innovations”, he offers three valuable insights: what qualities make the
innovation interesting to population, the importance of communication among peers for, and
undertanding of the needs of five different segments of adopters. To begin with, let us discuss the
process of individual adoption (see Figure 3).
Figure 3. Innovation-decision process. Source: Rogers (1962)
At knowledge step, a person sees the innovation, but does not know a lot about it. Moreover, he is
not willing to find out more. However, due to instant information flow about innovation, at
persuasion step, the individual becomes interested and starts seeking the information about the
innovation. Here play the major role five factors, that affect the individual’s perception of the
innovation: relative advantage, compatibility, complexity, trialability, and observability (see Table
1). At decision step, the individual weights pros and cons of the innovation and make a decision
whether he adopts it. Since it is hard, or even impossible, to predict individual behavior, it is hard to
get any empirical evidence on this stage. If the person acceps the innovation, he moves on to the
implementaion step, at which he individual uses the innovation and evaluates its usefullnes. If the
individual considers the innovation still useful, he finalize his decision to use the innovation at
confirmation step.
14
Table 1. Five main factors that influence the innovation adopters. Source: Rogers (1962)
Factor
Describtion
Relative advantage
The benefits of adopting the new technology compared to the costs
and in relation to other alternatives
Compatibility
The extent to which adopting and using the innovation is based on
existing ways of doing things and standard cultural norm
Complexity
The difficulty involved in using the new product
Trialability
The extent to which a new product can be tried ona limited basis
Observability
The extent that benefits of the new product are observable to everyone
We have discussed the process of adoption – the stages that the individual undergoes to
adopt the innovation; now let us discuss the process of diffusion – the group phenomenon that
describes how the innovation spreads among whole population. According to the diffusion theory, a
population can be broken down into five different segments, based on their willingness to adopt the
innovation: innovators, early adopters, early majority and laggards (see Figure 4). For the innovation
to be successfully adopted, it should address the needs of these five successive segments. Robinson
(2009) explains the characteristics of these segments and provides ways to reach them:
15
Figure 4. The diffusion of innovations according to Rogers. Each adopters’ category accounts for a
certain percentage of the market. Source: Rogers (1962)
Innovators are people who love innovation for its newness. They are willing to take risks and
have sufficient financial resources to buy new innovative products at the time of their introduced,
even if these products have not proved their efficiency and workability. The innovator has to do
little, if anything, to capture this type of customers.
Early adopters love innovation for its efficiency. They usually are step ahead of their peers,
more successful and wealthy than their peers are. They also like discussing their success and be in
the spotlight. There are several ways to get early adopters. The innovator can offer face-to face
support, ask their opinion as experts about the product, reward their egos with media coverage and
promote them as fashion leaders.
Early majority are pragmatists who want to use innovation, but only if it has been approved
by other people who are similar to them. They are risk-averse and cost sensitive, so they value
proven, easy-to-use solution. The innovator should stimulate the buzz around product to reach early
majority: he can set up competitions among users, use mainstream advertisements that features
16
endorsements by similar folks. Moreover, he should make the product convenient and guarantee its
performance.
Late majority art pragmatists who hate risks and are not comfortable with new ideas and
products. They follow mainstream standards because they fear not to fit in. That is why the
innovator should promote not only product benefits, but also social norms that foster use of the new
product. Moreover, the innovator should emphasize that the-state-of-are product is free of any risks.
Laggards are people who see high in adopting a new product. They usually adopt product
only if there are no other alternatives. The innovator should restrict their influence on the
population.
Rogers’ book was a great advancement in the science; however, it was very literary and did
not include mathematical representation. Bass (1969) solved this problem in his article “A new
product growth for model consumer durables”, which became one of the most highly cited papers in
the marketing literature. He created Bass Diffusion Model, a simple equation that describes the
process of how new products are adopted in a population, to mathematically represent a life-cycle
sales curve of innovative product over time.
𝑎 (𝑡) = 𝑀𝑝 + [𝑞 − 𝑝]𝐴(𝑡) −
𝑞
𝑀
𝐴2 (𝑡),
(1)
where
a(t) - adopters (or adoptions) at t (measured in years);
A(t) - cumulative adopters (or adoptions) at t;
M - the potential market (the ultimate number of adopters);
p - coefficient of innovation;
q - coefficient of imitation.
The adoption curve is fully defined by M, p, and q. M can be obtained by market definition
and analysis. P and q can be obtained from two sources: analysis of historical data for similar
products or regression analysis based on actual sales of the new product during the beginning of the
cycle. If neither is available, you can use average (for different historical products) p and q
coefficients: p = 0.003 and q = 0.38. When M, p and q are obtained, the equation predicts future
sales of the product. For instance, the potential total market for on-line streaming services in Russia
(M) and predicted sales in subsequent years were calcualated, using standard coefficients (p, q) (see
Figure 5). Knowing the percentage of the market covered in a particular year, it can be predicted
17
when each type of adopters start adopting, so the company can adjust the product’s marketing
strategy to the needs of the “current” segment. For example, from 2015 to 2019, it is time for early
adopters, who make up 13.5% of the total market right after innovators who make up 2.5%, so
during this time the on-line steaming company should offer face-to face support, reward early
adopters’ egos with media coverage and promote early adopters as fashion leaders. These actions
help to make the adoption of the product successful.
New users
Total users
50,000,000
45,000,000
40,000,000
People
35,000,000
30,000,000
25,000,000
20,000,000
15,000,000
10,000,000
5,000,000
0
2010
2015
2020
2025
2030
2035
Year
Figure 5. Curve of adoption of on-line streaming services by Russian market
Bass model is widely used in marketing. However, you should keep in mind several caveats
while using it. First, it accounts for the adoption of the product, not a particular brand: it may predict
the adoption of the mobile phone, but not that of Nokia. Second, q and p coefficients vary across
geographic locations. Van den Bulte (2005) argues that the average coefficient of innovation p in
Europe and Asia is roughly half of that in the U. S, and the average coefficient of imitation q in Asia
is roughly a quarter less than that in the U. S. and Europe. Neglecting this variation may cause
severe mistakes in the predictions.
Some scientists believe that there is a “chasm” between early adopters and the early majority
users in a product’s life cycle (see Figure 6) into which many promising products fall, unable to
make the leap. This situation happens because early adopters and early majority value different
things in the products and because representatives of different segments generally look at their peers
18
as references, rather than at representatives of the other segment. In order ot overcome the chasm,
an innovatior should target a specific niche, “beachhead”, from which he can easier expand into the
whole market (Moore 1991).
Figure 6. The chasm. Source: Moore (1991)
It is interesting to mention that the creator of diffusion theory, Rogers, denies the chasm.
Nethertheless, the existence to the chasm was proven by several researches, one of which is a
research by Chandrasekaran and Tellis, conducted in 2011 to study the other related phenomenon
called saddle. A saddle is a “phenomenon characterized by a sudden, sustained, and deep drop in
sales of a new product, after a period of rapid growth following takeoff, followed by a gradual
recovery to the former peak ”. One type of this phenomenon is presented on the Figure 7.
19
Figure 7. Saddle caused by chasm. Source: Chandrasekaran and Tellis (2011)
The authors collected time series data on sales and market shares of six kitchen/laundary and
four infromation/entertainment products in 16 European countries from 1950 to 2008. After the
analysis they found out that saddle is pervasive across countries. They also found out that the cause
of saddle is different for kitchen/laundary and infromation/entertainment product categories: for
regular products, such as kitchen/laundary products, the explanation is covered in business and
technological cycles, whereas for innovative products, such as infromation/entertainment products,
the explanation is covered in differences among adopter segments, i. e. in chasm, proving that chasm
actually exists.
Not only sociologists and mathematics cared how technologies are accepted by population,
but also information systems professors. Davis (1989) introduced the technology acceptance
model (TAM), an information systems theory that models how users come to accept and use a
technology (see Figure 8). The model emphasize two factors that highly affect the acceptance of the
new technology: perceived usefulness – "the degree to which a person believes that using a
particular system would enhance his or her job performance", and perceived ease-of-use - "the
degree to which a person believes that using a particular system would be free from effort". It is
important to mention that not actual usefulness and ease of use matters, but perceived one. Davis
found a correlation of 0,63 between perceived usefulness and actual use, and correlation of 0,45
20
between perceived ease of use and actual use. In addition to establishment of the basic model, David
is credited for development of questionnaires for perceived usefulness and perceived ease of use
measurement, an instrument that exhibited validity and reliability.
Figure 8. The Technology Acceptance Model. Source: Davis, Bagozzi and Warshaw (1989)
Disruptive innovation is one more important concept that is used to describe the innovation
diffusion process. Clayton Christensen (1997) defines disruptive innovation as a “process by which
a product or service takes root initially in simple applications at the bottom of a market and then
relentlessly moves up market, eventually displacing established competitors”. The established
companies pursue sustaining innovation for the high end of the markets because historically this
course of actions helped them to succeed (see Figure 9). However, at point “b”, customers stop
valuing enhancements in the product performance, since these enhancements are not necessary. For
example, there is no difference for a regular consumer whether his camera has 20 or 25 megapixels.
On the other hand, at the point “a”, disruptive innovation enters the low end of the market, where it
satisfies the market’s needs. At the point “c”, the disruptive innovation satisfies the needs of the
whole market, including the high end of the market. Afterwards, customers cannot see the difference
between incumbent and innovator’s product performance, whereas the innovator’s product is
cheaper, so the innovator’s product starts to dominate the market.
21
Figure 9. How disruptive innovation wins the market. Source: Christensen (1997)
Christensen provides several examples of disruptive innovations (disruptors) and technologies that
were crowded out by these innovations (disruptee) (see Table 2). He also argues that established
companies do not regard the disruptive technology as a serious business threat until it is too late,
since this technology is not mature enough to satisfy the high-end market on which the established
company is focused, so the company do not think that the new technology can get its customers. The
fact that the new technology wins the low-end market does not matter, since its profitability is too
low for established company to be interesting for the established company.
22
Table 2. Examples of disruptive innovations and incumbent technologies. Source:
Christensen (1997)
Disruptor
Disruptee
Personal computers
Mainframe and mini computers
Mini mills
Integrated steel mills
Cellular phones
Fixed line telephony
Community colleges
Four-year colleges
Discount retailers
Full-service department stores
Retail medical clinics
Traditional doctor’s offices
Paap and Katz (2004) provide an alternative explanation regarding how new technologies
substitute the old ones in “Anticipating disruptive innovation” article. They argue that technology
substitution occurs only when the current technology cannot address the unmet customers’ needs.
That is wrong to assume that technology comes first, since customers’ needs and technologies come
together. Guiding by this principle, the authors identified can identify three patterns of technological
substitution:
1. The old technology cannot effectively address the same need (Case 1).
2. The previous need matures, and a less important need becomes dominant, whereas the old
technology cannot effectively meet previously less important need (Case 2).
3. The environment changes, creating a new need (Case 3).
The first case is probably the most common pattern of technological substitution. This type
occurs when the current technology cannot longer meet the customers’ need, even if the type of the
need remains the same. For example, writeable DVD technology is replacing CD technology
because people want greater storage capacity from the disk. Another example is that wideband
technologies are replacing dial-up technologies for access to the Internet, since the customers
require higher speed. In both examples, customers’ need remains the same (capacity or speed), but
the customers want more of it (capacity or speed).
The second case comprised many examples of disruptive technologies. It occurs when
customers do not value any further enhancements of a product or service that address a particular
need, but their attention shifts to other needs that cannot be satisfied with the existing product.
However, incumbents often tend to continue improvement of the existing technology, even though
23
the customers do not value them so much anymore – this is the same situation that was described by
Christensen (1997). A historical event when 51⁄4-floppy disks substituted 31⁄2-inch disks is a good
illustration to this situation. The main customer need – storage capacity – remained the same for a
while. However, when this need reached its limit of 2.5 megabyte, existing, but less important needs
(disk size and durability) emerged to drive the future customer behavior, so people switched from
51⁄4-floppy disks to 31⁄2-inch disks.
The third case happens when a new need emerges – exactly new need, not an old, less
important need. The new need can appear because of changes in political, economic, social or
technological environment. The invention of a wash and wear fabric blend, a new fabric technology,
can serve as an illustration to this concept. This invention created a need for washing machines to
have a “cool down cycle,” which would optimize the performance of the new fabrics. Whirlpool, an
American multinational manufacturer and marketer of home appliances, forecasted this need and
designed washing machines, featuring a “cool down cycle”. These machines meet the new customer
need and, therefore, are very successful commercially.
1.3 Innovations in companies
A huge number of factors influences innovation success of organization. In order not to get
confused in their variety, it is advisable that we have a framework that categorize these factors. The
Organization for Economic Co-operation and Development in Oslo Manual (1997) suggests
“Innovation policy terrain” framework that distributes all these factors into four major categories:
innovation dynamo, transfer factors, science and engineering base and framework conditions (see
Figure 10). These categories concern business companies, science and technology institutions, and
issues of transfer and absorption of technology, knowledge and skills. In addition, the range of
opportunities for innovation is influenced by the surrounding environment of institutions, legal
arrangements, macroeconomic settings, and other conditions that exist regardless of any
considerations of innovation.
24
Figure 10. The innovation policy terrain - big picture view. Source: OECD (1997)
Each of four domains comprises a set of factors that are outlined on the Figure 11. The
components of three domains – framework conditions, science and engineering base, and transfer
factors - are thoroughly explained, whereas the components of the last one - innovation dynamo –
are not. That is because the scientific world have not already agreed on a particular model that
describes factors that shape innovation at the firm level. As Oslo Manual explains, “Many attempts
have been made to construct models to shed light on the way innovation is generated within firms
and how it is influenced by what goes on outside firms... However, some serious question marks
hang over all the available models”.
25
Figure 11. The innovation policy terrain - detailed view. Source: (OECD 2005)
Since the 1950s, there has been a proliferation of innovation models, each purporting to
explain and/or guide the process of innovation within industrial firms. Rothwell (1994) analysed of
state-of-the-art models of innovation processes at the firm level and classified these models into five
generations in his article “Towards the Fifth‐generation Innovation Process”.
First Generation Models: Technology Push (1950s–Mid 1960s)
These models, so called technology push models, were simple and linear (see Figure 12).
They were developed in 1950s and consider innovation successive process that begins from R&D
stage. Because these models put significant emphasis on R&D, companies and governments used
them to justify additional R&D spending to stimulate innovativeness and, in turn, business growth.
26
Figure 12. First generation technology push models. Source: Rothwell (1994)
Second Generation: Demand Pull Models (Mid 1960s–1970s)
Demand pull models were also simple and linear (see Figure 13). They were different from
technology push models because they implied that innovation process started not from R&D, but
from customer needs. However, R&D was the next step, since it realized spoted market needs.
Figure 13. Second-generation demand pull models. Source: Rothwell (1994)
First and second-generation innovation models were widely criticized for several reasons.
First, unlike models, innovation process in real companies is usually non-linear. Many innovation
activities happen simultaneously, not one by one, and there is much more feedback among the stages
that the models suggest. For example, product prototype may be returned to design department for
re-design if a marketing department think that customers will not buy such a product. Moreover,
sometimes innovation process is chaotic, especially in the early stages when a new product concept
is being generated and tested. Second, the models neglect to consider the influence of external
factors (environment, suppliers, customers, etc.) on the company’s innovation performance.
Participation in technological conferences, university lectures, exhibition, as well as discussion with
buyers and supplies contribute significantly to the innovation process. Third, the models do not
outline what happens on each stage in details – only very big-picture sequence of activities is given.
Forth, the innovation process is regarded as rigid, allowing no variation for the companies in their
ways to be innovative.
Third Generation: Coupling or Interactive Models (1970s)
Coupling or interactive models illustate the idea of interaction between science and
technology and the market (see Figure 14). The process of interaction comprise complex
communication paths both within and outside the organization; this process may not be continuous
but discrete. Rotwell also noted, that unlike the first and second generation models, third generation
27
models explicitly connect the marketplace, science and technology community and the decision
making of the firms.
Figure 14. Third generation coupling or interactive model of innovation. Source: Rothwell (1994)
The authors of third generation coupling models made a great advancement, addressing
weaknesses of the previous models. For example, paying attention to external environment, these
new models take into account feedback from the diffusion stage and companies’ need to improve its
product quality and decrease costs to overcome competition. Nevertheless, they do not incorporate
external factors, such as country’s legal environment or technology regulations, sufficiently enough.
Fourth Generation: Integrated Models (1980s)
Compared to third generation models, forth, so called integrated models, incorporated more
and more feedback loops and communication (see Figure 15). Japanies automobile companies
during 1980s led the scientist to the development of the new models, which included considerable
functional overlaps between activities of different departments, as well as companies’ external
integration with activities in other companies including suppliers, customers and, in some cases,
universities and government agencies.
28
Figure 15. Fourth Generation: Integrated Models (1980s). Source: Rothwell (1994)
Fifth Generation Systems Integration and Networking Models (Post 1990)
Fifth generation systems integration and networking models underscored that innovation
relied on learning and networking, both of them within and between firms. Partnerships among
companies, joint ventures and corporate alliances, which were popular during 1980s and 1990s,
directed researchers towards models that emphasised vertical relations, such as strategic alliances
with customers and suppliers, and collaborating competitors. Rotwell argues that time pressure on
leading edge innovations also affected the development of fifth generation models: in order to
increase new product development speed and efficiency, the models suggested to use complex IT
tools. According ot Rothwell, use of complex electronic tools that operates in real time and
automates the innovation process within the company is actually a defferentiating factor between
forth and fifth generation models. He also holded that cost and difficulties associated with
intoduction of complex IT solution were offsetted by obtained benefits, such as speed of innovation
and attaining market leadership.
29
Figure 16. An example of a fifth generation systems integration and networking model. Source:
Trott (1998), cited in Hobday (2005)
However, there is little evidence that companies have adopted fifth generation models of
innovation and that increased use of IT benefits compensate its drawbacks. Some studies showed the
negative sides of IT, such as difficulties of adoption by employees, and high up-front costs to set up
the system. They also emphasized that organizations need to be prepared to implement electronic
tools for complex topics such as innovation, otherwise IT may hamper the company’s performance.
Moreover, the usefulness of IT solution in innovation process depends on the nature of the product
and technology in question. While electronic tools may support lower level tasks, they are unlikely
to substitute human interactions, team building, group work and the leadership required for complex
tasks.
Not only criticized particular generations of models, but also all of them for several reasons.
First, Mahdi points out that the models usually provide one best way of innovation process,
eliminating alternative paths (2002, cited in Hobday (2005)). That is not a bad thing, but many
companies tend to revert their innovation process to the simple stages that were proposed in the
models, and this simplification does not always affects business positively. Mahdi argues that
evidence demonstrates that major differences in innovation process exist within and across different
30
industries. Moreover, these differences persist over time and are not a deviation from a norm. That
is, any generalized model of innovation process is misleading, as innovation process is determined
by a set of historical and external factors. To illustrate his ideas, Mahdi brings the example of
software industry, in which software development process usually goes iteratively, “first a rough
specification of the software requirements is made, then a prototype is developed that is then tested
and modified”.
Let us have a look at Stage-Gate model by Cooper (2008). According to this model, product
innovation starts with an idea that goes through severlas stages and gates towards its market launch
(see Figure 17). The project team works on the product at stages, while the management team makes
decision whether product moves on to the next stage at gates. The Stage-Gate model is extremely
popular in business: 85% of North American companies use it for product development, increasing
efficiency and decreasing cost of development. Nevertheless, such a stict process limits innovation
in certain cases, since innovation may be spontaneous and creative process that cannot be fully
formalized.
Figure 17. Stage-gate model. Source: Cooper (2008)
Second, most of the models assume that the companies behave too rationally, being able to
hypothesize a solution to an innovation problem, such as a new product development, and then
systematically solve the problem, using a standard toolkit such as design thinking, prototype testing
and market research. However, this case is only possible when companies have enough experience
to make an educated guess about the potential solution, otherwise, they have to use iterative
approach, experimenting, making mistakes, and trying again. Therefore, the models in question may
be applicable only to experience firms that have enough experience in the field.
31
Third, the models lack a coherent theoretical base, which is important because it can help to
put innovation within the wider organizational and strategic context in which it belongs. Indeed, the
state-of-the-art models treat innovation is separate process, failing to understand that it is tightly
connected with the firm’s strategy, culture and capabilities. This idea is upheld by empirical
evidence (Hobday 2005).
Forth, all the models deal with innovation leaders, neglecting latecomers. This fact is
especially important for scientist who research innovation in developing countries, e.g. Russia, since
many companies there do not develop innovations themselves, but rather adopt them from abroad.
Therefore, the models for developing countries should be different from those for developed ones.
Fifth, the majority of models deals describe processes in the large corporations, not paying
attention to medium and small companies, where innovation process usually do not have any formal
stages and domain. For instance, a small company may not even have R&D department. Therefore,
for such companies it is definitely inappropriate to apply existing models.
In order to address the issues with previous models, an attributive model of innovation was
created. This model outlines what qualities the company should have to be successful in innovation.
The main difference with the previous models is that the attributive model does not aim to depict
any particular process that companies should follow, but rather identify key attributes that the
company should have to be successful in innovation. The analogy can be a good illustration for the
new model ides. Both Figure 18 and Figure 19 present the models of a successful exam. However,
they use a different approach to model the phenomenon. Figure 18 depicts a sequence of actions that
a student should do to get a good grade. This sequence resembles the-state-of-art innovation models,
especially first and second generation. On the contrary, Figure 19 does not say a word about the
optimal process, but points out key components of a successful exam: knowledge, student’s
condition at the exam day, and his relationship with the professor. Needless to say, the later model is
much more practical the former one.
Study hard during
the semester
Come to the exam room
at the exam time
Answer correctly the
questions
Give your paper to the
professor
Figure 18. Illustration of the new model idea: a sequence of action to pass an exam successfully
32
Figure 19. Illustration of the new model idea: determinants of a successful exam
Let us come back to the innovations in Russian companies. The attributive model of
innovation is presented on the
Figure 20. According to the model, a set of monetary and non-monetary determinants is
correlated with the innovation performance of the company. All these determinants, except
“Marketing preparation for product innovation”, are relevant to product and process type of
innovation.
33
Figure 20. Conceptual model of determinants of product innovation performance. The model for process innovation performance is
the same, except it does not have “Marketing preparations for product innovations” element.
34
How AMI addresses issues of the previous models? First, unlike previous generations of
models, AMI does not provide one best way of innovation process, eliminating alternative paths,
but rather enlist drivers that make the company successful in innovation. In AMI model in does
not matter whether, for instance, market need or technological development comes first, since its
non-sequential nature. What matters is that the company does have to pay attention both to
market needs and technology development, and this idea is addressed in the model (in
“competitive and market intelligence” branch).
Second, non-sequential nature of the model eliminates issues with hype-rationality,
because non-sequential nature allows for trial-and-error experimentation and rudimentary
activities.
Third, AMI model rests on a solid theory, namely modern resource-based theory. Teece,
Pisano and Shuen (1997) in their article “Dynamic Capabilities and Strategic Management”
highlights three domains that afferct the company’s competitive advantage – the company’s
processes, positions and paths. The authors define these components as follows:
By managerial and organizational processes, we refer to the way things are done
in the firm, or what might be referred to as its routines, or patterns of current
practice and learning. By position we refer to its current specific endowments of
technology, intellectual property, complementary assets, customer base, and its
external relations with suppliers and complementors. By paths we refer to the
strategic alternatives avail- able to the firm, and the presence or absence of
increasing returns and attendant path dependencies.
The AMI model’s determinants represent positions and processes as sources of
competitive advantage (see Table 3).
35
Table 3. The connection between AMI model and competitive advantage elements
Resource-based theory
AMI model
Processes
Innovation strategy of the company, competitive and market
intelligence, and ideas management
Positions
The amount of expenditures of innovation
Paths
Forth, AMI model fits both innovation leaders and latecomers, and both large
corporations and small companies, because the determinants of AMI neglect organizational
structure and encompass sources of innovations far beyond only R&D department. If the
company acquire external inventions, or just copycat the technology using competitive
intelligence, the AMI model still works.
1.4 Justification of AMI model determinants
This group of determinants is based on Oslo Manual taxonomy (OECD 2005). However,
the ideas from Olso Manual, which were created for studies in developed countries, were
simplified and adjusted to Russian realms and companies’ structures, making it possible to apply
general ideas of Oslo Manual for studies in Russia, which is not as developed as OECD countries
are. From this point onwards, here are presented not the original Oslo Manual ideas, but the
adjusted ones.
Research and experimental development (R&D) comprises creative work undertaken on a
systematic basis in order to increase the stock of technical knowledge in the company. Acs and
Audretsch (1988) created and tested a model that suggests that innovation performance is
influenced by R&D expenditures. The results showed that innovation performance is influenced
by R&D expenditures. Crepon, Duguetb and Mairessec (2006) studied the links between
productivity, innovation and research at the firm level. They found that innovation output of the
company increases with its research effort. Thus, it is assumed that expenditures on R&D are
positively correlated with product and process innovation.
Hypothesis 1a: There is a positive relationship between expenditures on R&D and
product innovation.
36
Hypothesis 1b: There is a positive relationship between expenditures on R&D and
process innovation.
R&D is not the only way for the company to obtain new technologies and know-how.
The company may also purchase patens, inventions and know-how from the other companies,
external research institution, abroad and so on. The idea of getting knowledge from outside is
highly appreciated by Chesbrough (2003) in his book “Open innovation: The new imperative for
creating and profiting from technology”, which set up a new trend in innovation studies and is
widely cited in different papers on innovation. Chesbrough argues that companies should seek
the opportunities to get external innovative ideas into practice to be successful in innovation. For
instance, Suzlon and Goldwin, India and China’s leading wind turbine manufacturers, acquired
licenses for technologies to produce wind turbine (Lewis 2007). Acquisition of external
knowledge enhance innovation productivity, as it broadens the company’s knowledge that
company can use in product development and operations. Moreover, it increases the company’s
understanding of the market and technological trends (Yli-Renko, Autio and Sapienza 2001).
Maurer (2010) found a correlation (p ⩽ 0.001) between knowledge acquisition and product
innovation. Thus, it is assumed the following:
Hypothesis 2a: There is a positive relationship between expenditures on acquisition of
external knowledge and product innovation.
Hypothesis 2b: There is a positive relationship between expenditures on acquisition of
external knowledge and process innovation.
Besides spending on R&D and external knowledge, there one more way to facilitate
innovation in the company. The company can buy “capital goods, both those with improved
technological performance and those with no improvement in technological performance [, but]
that are required for the implementation of new or improved products or processes (OECD
2005)”. This category includes machinery, instruments, equipment, computer software, and other
capital goods. The up-to-date infrastructure and equipment are particularly important in case of
Russian companies, many of which work with outdated infrastructure and very old equipment.
For instance, at the age of 3D printers, some industrial companies operate on 70-year-old
equipment obtained from after-war Germany. Given this condition, the main concern of the
managers is not how to innovate more, but how to make these machine tools not break down.
37
Opposite, if the company uses modern equipment, the managers have time to think about
innovation. Thus, it is assumed the following:
Hypothesis 3a: There is a positive relationship between expenditures on acquisition of
machinery, equipment and other capital goods and product innovation.
Hypothesis 3b: There is a positive relationship between expenditures on acquisition of
machinery, equipment and other capital goods and process innovation.
We have talked about physical objects so far. No let us switch to the employees. Now
much company invests in their education is important for its innovation success. However, in
case of developing countries, we have a caveat about the effect on training on innovations. In
order to understand it, we should go to the theories proposed by Utterback and Abernathy (1975)
and Kim (1980).
Utterback and Abernathy (1975) found out the relationship between the type of
innovation, the competition type and technological processes development level. As it presented
on Figure 21. Innovation and stage of development. Source: Utterback and Abernathy
(1975)Error! Reference source not found., the more developed technological process is, the
less product innovation the company has, and vice versa. This situation occurs because of the
nature of competition for innovative products: at the beginning, companies compete based on
product variety, but at the end – based on production process efficiency. It is important for the
production processes to be flexible at the beginning, since the company experiments with
products to find the type of product that customers like most, so the processes should be easily
adjusted to the ever-changing products. Even if some technological processes improvements
happen at this stage, they tend to be rare and simple. On the other hand, then the companies
finally understand the product that customers value most, the locus of competition moves to the
production: the companies want to save money on production, improving technological
processes through process innovation.
38
Figure 21. Innovation and stage of development. Source: Utterback and Abernathy (1975)
Utterback and Abernathy studied developed countries. The attempt to understand the
process of innovation in developing countries was made by Kim (1980) in his article “Stages of
development of industrial technology in a less developed country: a model”. The author
suggested a three-stage model, in which developing countries moving from acquisition of foreign
technology, to assimilation and eventually to improvement. First, the companies buy developed
foreign technologies that include packaged assembly processes that require only very limited
interventions from the buyer. Second, the companies acquire not the new technologies itself, but
technologies how to develop processes and design new technologies. Third, companies start
producing new and innovative products. To summarize, the sequence of events is opposite of that
of developed countries. The connection between these two sequences was depicted in the work
of Lee, Bae and Choi (1988) (see Figure 22).
39
Figure 22. Innovation process in developed (the top of the picture) and developing (the bottom
of the picture) countries. Source: Lee, Bae and Choi (1988)
According to and Abernathy (1975) and Kim (1980), the process of innovation starts with
process innovation in developing countries, and the process innovation may hamper product
innovation, since in order to be effective, the processes have to be rigid and hard to change. Like
the adoption of foreign technologies, that enables producing something more efficiently, but in a
standardized and rigid way, training of employees teaches them to work efficiently, using
standardized techniques and technologies, making them more creative in process domain, but
40
less creative in product domain. A training usually makes employees focus on how to make the
production process more efficient; however, since the attention of employees is limited, and they
start paying more attention to production processed, they also start paying less attention to
product improvements. This attention switch as well as increased process rigidness increase the
level of process innovation in the company, but decrease the level of product innovation. Thus, it
is assumed the following:
Hypothesis 4a: There is a negative relationship between expenditures on employee
training and product innovation.
Hypothesis 4b: There is a positive relationship between expenditures on employee
training and process innovation.
Product innovation occurs when market needs are coupled with technological
developments (Paap and Katz 2004), since no matter how good the product is, somebody has to
buy, if the company wants to be successful. In order to convince the customers to buy an
innovative product, the company has to invest in market preparation for the innovative product.
According to Oslo Manual, “market preparation for product innovations can include preliminary
market research, market tests and launch advertising for new or significantly improved goods
and services (OECD 2005).” These activities are designed to make the new product sellable, the
condition required to make the innovation not only created, but also implemented. The product’s
success of failure heavily depends on the quality of marketing managers product launch
preparation (Bearda and Easingwood 1996). Bonnin, Segard and Vialle (2005) argue that in
order to ensure the company’s long-term surival, the company should be able to successfully
introduce new products into marketplace. This capability is even more importnant today, since
the rate of technological and customer needs changes has risen drastically. Thus, it is assumed
the following:
Hypothesis 5a: There is a positive relationship between expenditures on marketing
preparations for product innovation and product innovation.
There is no 5b hypothesis, since in process innovation company does not sell anything.
It is important to make every employee understand how important innovation is to make
the company successful in innovation. Everyone from CEO to the worker at the factory or in the
41
office should realize the importance of innovation and try to be innovative. In the article “Five
Ways to Make Your Company More Innovative”, Kanter (2012) wrote, “Put innovation at the
heart of strategy, and tout it in every message (Emmons, Hanna and Thompson 2012)”.
For company to be successful in innovations, the company should not only carry out
innovation activities, but also integrate them under the umbrella of a single innovation strategy.
Cassiman and Reinhilde (2002) provided empirical evidence of this idea, using data on Belgian
manufacturing companies. The company should clearly define its strategic goals in order to
achieve them efficiently. Companies that fail to articulate their innovation strategy clearly often
have difficulties with everyday operations because innovation teams scruple from one
opportunity to another, being not concentrated on a single defined goal.
Whereas employees of companies abroad usually take pains to be innovative, Russian
employees may concerned with more basic things, such as overcoming problems with public
officers, inspections or fixing broken printer. For example, as far back as in 1989, Japanese
companies got on average 38 ideas per employee, despite the average reward for idea was set to
be $2.83 (Gupta and Tyagi 2008). Since $2.83 were not big money, the employees were
motivated by other factors. No matter what exact factors motivate them to innovate; they would
not have achieved such a good performance, unless they had realized how important innovation
was for their company. At the same time, in Russian environment, in which employees face a
great deal of unexpected problems and administrative obstacles, it is necessary that employees
instantly be reminded of the importance of innovativeness for the company’s well-being. It order
to be innovative, the company should have points regarding innovation in its strategy,
communicate them to its employees and make them believe that innovation is important. Thus, it
is assumed the following:
Hypothesis 6a: There is a positive relationship between features of innovation strategy
and product innovation.
Hypothesis 6b: There is a positive relationship between features of innovation strategy
and process innovation.
According to Oslo Manual, something new is only considered innovation, if it is
implemented (OECD 2005). For product innovation, it means that the innovative product should
42
be welcomed by the customers. In the second section of this thesis, “Diffusion of innovation”,
was said a lot about how customers adopt a new product and what the company should do to
facilitate the diffusion process, so these points will not be discussed again here. It will be just
pointed out that in order to achieve market success, the company has to perform marketing
intelligence.
Many innovations failed because the companies did not pay enough attention to customer
needs and external environment. Cinerama, the very first widescreen projection format and
prototype of IMAX, is one of the examples of such a failure. Projecting a Cinerama film meant
projecting three synchronized 35mm projectors simultaneously onto a gigantic curved screen.
The results were visually astounding and far ahead of any other method of the time. The
drawback of this technology was its price and complexity to execute. Three projectionists had to
project the film from three projectors synchronously that requires projectionists to be extremely
high skilled. Since it was hard for cinemas to find so skilled projectionists and Cinerama
technology was so expensive, only few theaters were willing to pay for this technology. As a
result, only a couple of dozen films ever used the Cinerama format (Floorwalker 2013). Thus, if
the company fails to understand its customers, the innovative product may fail to be adopted,
even if it is amazing.
One more important activity the company should be involved in to be successful with
innovations, and, particularly, with disruptive innovations is competitive intelligence. While
market intelligence is used to gather data on external environment, competitive intelligence as a
process of gathering actionable information on competitive environment. Entrepreneur magazine
describes competitive intelligence as follows:
Competitive intelligence essentially means understanding and learning what is
happening in the world outside your business so you can be as competitive as
possible. It means learning as much as possible - as soon as possible - about your
industry in general, your competitors, or even your county's particular zoning
rules. In short, it empowers you to anticipate and face challenges head on.
Competitive intelligence is different from marketing intelligence in its proactive nature: it
looks for weak signals in external environment to predict market trends and competitors’ actions.
43
Gilad, one of competitive intelligence guru, explains that competitive intelligence is a three-step
procedure - risk identification, competitive monitoring, and management actions (2003).
First step is risk identification with war games. During this stage, competitive intelligence
specialists have to do the following:
1. identify key drivers in the industry, assess their probability and create scenarios;
2. obtain the information about the competitors;
3. understand strengths and weaknesses of competitors and their company;
4. identify blind sport of the competitors’ and our company management;
5. play war game to assess the potential competitors’ reaction to the changes it the industry.
After that stage, competitive intelligence specialist will understand the areas of major
risks - e.g. mass introduction of a new production technology that boost production process
efficiency, but so far is too expensive to implement - competitive intelligence specialist can find
the indicators that the company should follow afterwards to understand when the company
should react to the risk. In case of the new production technology the significant drop in
implementation price may serve as such indicator, since when it occurs, the competitors may
implement this technology faster than your company do, gaining competitive advantage over
your company.
Second step is competitive monitoring. During this stage, the company should monitor
the environment if any risk is carried out. There should be a team responsible for the monitoring.
The team should include interior and exterior experts. Interior experts should have access to the
information they need to monitor. They should be able to understand deeply the information and
face this information during their regular work, not only for monitoring reasons. That helps to
avoid superficial analysis. External experts should be included in the monitoring team, as their
vision is not distorted by work in particular company and as they may know some information
that employees of the particular company do not know.
Third step is management actions. It is important to mention that competitive intelligence
should only deliver insights to management and be persuasive, not to implement created
recommendations that is management responsibility. At this step, Gilad recommends that
competitive intelligence specialists give conclusions and preliminary plans of actions to
44
management rather than raw data. That is because competitive intelligence specialists know the
data better than managers, who did not collect the data, and because managers sometimes use
raw data not for the analysis, but for support of their opinion that they have already had. In case
of the introduction of a new production technology, which was used as an example of previous
stages, the managers have to implement this technology rapidly in their business in order not to
lose to the competition. Since marketing and competitive intelligence features are important for
the company’s innovation success, it is assumed the following:
Hypothesis 7a: There is a positive relationship between features of market and business
intelligence and product innovation.
Hypothesis 7b: There is a positive relationship between features of market and business
intelligence and process innovation.
Idea generation should not be treated as if it was only R&D business. Every employee in
the company, of even entities outside the company may help the company to be innovative.
Let us start from idea generation within the company. As it was previously mentioned, as
far back as in 1989, Japanese companies got on average 38 ideas per employee (Gupta and Tyagi
2008). This example illustrates how creative employees can be. In Russia, unfortunately, culture
of many companies does not facilitate idea generation: neither people are motivated to share their
ideas, not the formal idea collection processes exist. However, there are some exceptions from
this defective practice. For instance, Sibur, Russian largest integrated gas processing and
petrochemicals company, has an established process of collecting ideas from its employees.
There are special boxes at Sibur’s plants, in which employees can put notes, including
anonymous notes, with their ideas. Every week, this box is emptied and the notes are put on the
board, on which everyone can write what he thinks about the proposed ideas. If any idea is
upheld by the colleagues, this idea goes to managers. If the managers like the idea, it is
implemented and the author is paid. The company should leverage the opportunity to benefit
from its employees’ idea. Kanter (2012) explains how Verizon, the largest wireless
telecommunications provider in the United State, benefits from ideas of its employees from all
levels of company’s hierarchy:
45
Think of innovation strategy as a pyramid: big bets at the top, a few projects in
development in the middle, and a broad base of continuous improvements,
incremental contributions, and early-stage new ideas at the bottom. For example,
Verizon has placed big bets on Google's Android for smartphones and on fiber
optics for landlines, and now is seeking new ways that wireless networks could
run everything, including cars and refrigerators. It has projects in development
with GM's OnStar and in cloud computing. In addition, Verizon CEO Lowell
McAdam sees small "pots of gold" everywhere in the business, even in the
traditional landline side, preaching process innovations to technicians.
Not only employees can generate useful ideas, but also external parties. Chesbrough
(2003) created a special term for this situaion – open innovation, that is "a paradigm that assumes
that firms can and should use external ideas as well as internal ideas, and internal and external
paths to market, as the firms look to advance their technology”. He argues that companies cannot
remain competitive if they rely only on centralized internal R&D processes, since a number of
"erosion factors" have changed the business environment. This idea rests on several principles,
but for the sake of concision only the most important principle will be restated in this research not all smart people work for one single company, so the company must seek knowledge from
external sources. For example, it can set up a competition among our customers, so they can tell
the company how it should improve its product. This way of obtaining knowledge is cheap and
effective.
Hippel (2006) in his book “Democratizing Innovation” goes even further. He speculates
why and how product users develop and freely reveal innovations they make by themselves and
how companies can capitalize on this phenomenon. Since standardized products do not fully
fullfil customer needs, up to 40% of users modify these products. The users have two options:
either to hire a specialist, who modifies the product, or does modification by themselves. People
usually prefer doing modification by themselves because this way they can avoid agency costs
and enloy the modification process. Furthermore, after the modification is over, they usually
share the result of their work with people around, since the sharing improves their reputation,
creates positive network effect and facilitate innovation diffusion. Because of the emergence this
new customer-led innovation, companies should shift their attention from designing of new
46
products to better commerzialization of lead users innovation. Hippel views are novel and
progressive, however, they are too radical, especially in Russian case. However, Chesbrough’s
views on open innovation suits Russian context to some extent. For this research, it is assumed
the following:
Hypothesis 8a: There is a positive relationship between features of the company’s idea
management and product innovation.
Hypothesis 8b: There is a positive relationship between features of the company’s idea
management and process innovation.
Based on literature review, eight hypotheses regarding the determinants of innovation
performance was put forth (see Table 4). “+” means that there is a positive relationship between
the determinant and the type of innovation. “-” means that there is a negative relationship
between the determinant and the type of innovation. The absence of the sign means that no
hypothesis about the relationship between the determinant and the type of innovation was
proposed.
Table 4. Stated hypotheses
Hyp.
#
1
2
3
4
5
6
7
8
Product
Process
innovation innovation
Research and experimental development
+
+
Acquisition of other external knowledge
+
+
Acquisition of machinery, equipment and other capital goods
+
+
Training
+
Marketing preparation for product innovations
+
Features of the company's innovation strategy
+
+
Features of the company's competitive and market intelligence
+
+
Features of the company's idea management
+
+
The conceptual model of relationship between product innovation performance and its
Determinants
determinants is presented in the following equation:
47
Product innovation performance = f (Research and experimental development,
Acquisition of other external knowledge, Acquisition of machinery, equipment and
other capital goods, Training, Marketing preparation for product innovations, (2)
Features of the company's innovation strategy, Features of the company's competitive
and market intelligence, Features of the company's idea management)
The conceptual model of relationship between process innovation performance
and its determinants is presented in the following equation:
Process innovation performance = f (Research and experimental development,
Acquisition of other external knowledge, Acquisition of machinery, equipment and
other capital goods, Training, Features of the company's innovation strategy, Features (3)
of the company's competitive and market intelligence, Features of the company's idea
management)
48
1.5 Methodology
The insights from the literature review was helpful to build the conceptual model.
However, the conceptual variables should be operationalized to be used in the regression
analysis. All the conceptual variables and their operational proxies are depicted on the Figure 23.
Two types on the company’s innovation performance types are measured in this research product innovation performance and process innovation performance. Their measurement and its
justification is provided below.
As a starting point for operationalization of the conceptual variables, the article by
Albaladejo and Romijn (2002) was used. This article studies innovation capabilities in small
electronics and software firm in southest Engand, since the article provides a solid operational
framework to study innovation capabilities in the companies. The authors were focused only on
product innovation, since this type of innovation prevailed in their sample. They used three
dependent variables to measure the company’s innovation performance: incidence of major
product innovation, the number of patents, and product innovation index.
First, incidence of major product innovation. This variable is a simple binary variable that
indicates whether a company had accomplished at least one major product innovation during the
3 years preceding the survey. Being binary, this variable can only capture the existence of
innovativeness, but cannot capture the degree of innovativeness. In order to make this research
able to capture the differences in innovativeness, binary variables were not used in this research.
49
Figure 23. Operational model of determinants of product innovation performance. The model for process innovation performance is
the same, except it does not have “Marketing preparations for product innovations” element.
50
Second, the number of patents that the company filed during a limited period of time. The
weakness of that variable is that many innovations that small firms come up with are never
patented, especially in small companies and if the speed of technological advance in the industry
is high. The expense and effort needed to apply for patent protection and to deal with patent
infringements may be beyond the firm’s limited capacity; the pace of technological advance may
be so fast that it is not considered worthwhile to pursue patenting. Moreover, some innovations
may not be so fundamentally new as to qualify for patenting. Number of patents is not used in
this research because of its this variable’s disadvantages.
Third, product innovation index. The product innovation index can be a way to get
around the drawbacks of the previous indicators to some extent. It is based on extensive
qualitative information about the extent and significance of each firm’s innovative outputs
generated within a certain period of time. The table (see Figure 24) was used to assign scores to
the company. For example, if the company had fundamentally new to the world innovation,
featuring low scientific involvement, the firm got four points for its innovativeness.
Figure 24. Scale used for product innovation index in survey by Albaladejo and Romijn (2002)
However, it is questionable why science-intensive innovations should be valued more
than not science-intensive innovations. Being scientific is not a goal of innovation. Roughly, the
goal of innovation is to help the company to win the competition in the ever-changing world. If
some new marketing gimmick can help the company to achieve this goal, it should be consider a
true innovation. The product innovation index is not used in this research because of this
variable’s disadvantages.
The good measure of the company’s product innovation performance is the amount of
revenue coming from innovative products. However, the amount of revenue from innovative
products depends not only on the company’s innovativeness, but also on the company size. In
51
order to eliminate the dependence on the company size, the amount of revenue from the
innovative products should be divided by the measure of the company’s size, for instance,
revenue.
This reasoning was also used in papers of famous scientists before. Cooper and Edgett
(2012) in their paper on benchmarking, argue that percent of revenue coming from new products
is a good measure of the company’s new product performance. The only difference a new
product and an innovation product is the degree of novelty. Thus, the percent of revenue from
innovative products can be also used as a proxy of the company’s product innovation
performance. Furthermore, this reasoning can be also applied to process innovation
measurement. A portion of savings initiated by process innovation implementation from the total
company’s revenue was decided to be used as a measure of the company’s process innovation
performance. Information about the dependent variables is presented in the Table 5.
There are two groups of determinants of the company’s innovation performance:
monetary and non-monetary. The justification of their measurement is presented below.
Monetary determinants pertain spending on different things in the companies. Therefore,
their measurement is relatively trivium task: the researcher should just find out how much money
the company spends on different things. However, since the company’s size influence the
spending, the amount of spending has to be divided by the measure of the company’s size. Total
company’s revenue was used for this role. Therefore, to measure monetary determinants, a
percent of sales that is spent for something was used.
Non-monetary determinants are the features of the company’s processes and policies.
Because of time and financial constraints of this research, no field studies in the companies were
conducted, but rather the employees were asked to answer sets of questions regarding particular
processes, and then their answers were transformed into the measure of the processes or policy
quality. The employees were asked to identify their level of agreement with certain statements,
such as “The innovation strategy is actively communicated to the employees”, using 1-5 Likert
scale. Then, the median of the scores was calculated to get scores for the quality of the
company’s certain processes or policy. For example, when the quality of the company's
competitive and market intelligence was measured, the respondents were asked to identify their
level on agreement to the following statements:
52
The employees monitor the market, competitors and new technologies
External entities monitor the market, competitors and new technologies
The data from the monitoring is analyzed, for instance, with the use of scenario
planning or benchmarking
The results of analyses influence the management decisions
If the respondent assigned 4 points for the first statement, 2 points for the second, 5
points for the third and 5 points for the forth, the quality of the company's competitive and
market intelligence got 4,5 scores, since the median of (4, 2, 5, 5) is 4,5. This method was
borrowed from the work on political science by Manheim and Rich (1995).
For your convenience, all dependent and independent variables with their discription are
presented in the Table 5.
53
Table 5. Variables in the regression model
#
Concept
Measurement
Variable
type
Name in SPSS
Dependent variables
1
Product innovations
Portion of revenue from innovative
products to the total from the total Numerical ProdRev
company’s revenue
2
Process innovations
Portion of savings initiated by process
innovation implementation from the Numerical ProcRev
total company’s revenue
3
4
Independent variables
Research and experimental
Total R&D expenditures / sales
Numerical ResandDev
development
Acquisition of other external Spending on acquiring of external
Numerical Patents
knowledge
knowledge / sales
Acquisition of machinery,
5 equipment and other capital
goods
6 Training
Market
and
other
7 preparation for product
innovations
Features of the company’s
8
innovation strategy
Features of competitive and
9
market intelligence
Features
of
ideas
10
management
Spending / sales
Numerical NewEquip
Total training expenditures / sales
Numerical Training
Spending / sales
Numerical Marketing
Median of responses measured with
Numerical StrategyMedian
Likert scale
Median of responses measured with
Numerical BIMedian
Likert scale
Median of responses measured with
Numerical IdeasMedian
Likert scale
54
The operational model of relationship between product innovation performance and its
determinants is presented in the following equation:
𝑃𝑟𝑜𝑑𝑅𝑒𝑣𝑖 = 𝛽0 + 𝛽1 ∗ 𝑅𝑒𝑠𝑎𝑛𝑑𝐷𝑒𝑣𝑖 + 𝛽2 ∗ 𝑃𝑎𝑡𝑒𝑛𝑡𝑠𝑖 + 𝛽3 ∗ 𝑁𝑒𝑤𝐸𝑞𝑢𝑖𝑝𝑖 + 𝛽4 ∗
𝑇𝑟𝑎𝑖𝑛𝑖𝑛𝑔𝑖 + 𝛽5 ∗ 𝑀𝑎𝑟𝑘𝑒𝑡𝑖𝑛𝑔𝑖 + 𝛽6 ∗ 𝑆𝑡𝑟𝑎𝑡𝑒𝑔𝑦𝑀𝑒𝑑𝑖𝑎𝑛𝑖 + 𝛽7 ∗ 𝐵𝐼𝑀𝑒𝑑𝑖𝑎𝑛𝑖 + 𝛽8 ∗
(4)
𝐼𝑑𝑒𝑎𝑠𝑀𝑒𝑑𝑖𝑎𝑛𝑖 + 𝜀𝑖
)
The operational model of relationship between process innovation performance and its
determinants is presented in the following equation:
𝑃𝑟𝑜𝑐𝑅𝑒𝑣𝑖 = 𝛽0 + 𝛽1 ∗ 𝑅𝑒𝑠𝑎𝑛𝑑𝐷𝑒𝑣𝑖 + 𝛽2 ∗ 𝑃𝑎𝑡𝑒𝑛𝑡𝑠𝑖 + 𝛽3 ∗ 𝑁𝑒𝑤𝐸𝑞𝑢𝑖𝑝𝑖 + 𝛽4 ∗
𝑇𝑟𝑎𝑖𝑛𝑖𝑛𝑔𝑖 + 𝛽5 ∗ 𝑆𝑡𝑟𝑎𝑡𝑒𝑔𝑦𝑀𝑒𝑑𝑖𝑎𝑛𝑖 + 𝛽6 ∗ 𝐵𝐼𝑀𝑒𝑑𝑖𝑎𝑛𝑖 + 𝛽7 ∗ 𝐼𝑑𝑒𝑎𝑠𝑀𝑒𝑑𝑖𝑎𝑛𝑖 + 𝜀𝑖
(5)
The data that needed for this research were not available in any database, so primary data
had to be collected. Electronic questionnaire, created using Google Forms, were used to obtain
the data. Several ways to attract respondents to answer the questionnaire were used: first, the
researcher asked 600 acquaintance to participate; second, he publicized the need for cases in
GSOM alumni Facebook group (https://www.facebook.com/alumni.gsom); third, he publicized
the need for cases several publics connected to innovations on vk.com (https://vk.com/sciseek,
https://vk.com/public9464801,
https://vk.com/innovationnews,
https://vk.com/innovative_thoughts).
In order to make the questionnaire more understandable and to uncover any potential
problems, five people tested it before the launch. After they completed the questionnaire, they
were asked for feedback, which was analyzed and used to make final adjustments. The
questionnaire was strived to be as short as possible because if the questionnaire is long, people
often either do not participate in the survey at all or answer the last questions at random, since
the people get tired and bored by the length of the questionnaire. The final version of the
questionnaire was five pages long and could be answered in less than ten minutes.
The questions in the questionnaire can be divided into three group: general questions
about respondents and companies they represented, questions about financial ratios, and
questions about the company’s processes. General questions sought to gain brief understanding
of the person who took part in the survey and of the company that he represented (see Figure 30).
Questions about financial ratios seek to understand how much company spent on something or
55
got from something (see Figure 31). In order to make it easier for people to answer these
questions, intervals was used instead of just blank space in which people could white their
estimate. The range of possible answers was created based on how much the companies usually
spend on such activities. Nevertheless, if the respondent’s company was out of the range, the
answer choice “other” with blank space for the response was also available. The questions about
the company’s processes seek to understand the several processes in the company (see
Figure 32). Several statements were grouped in blocks on the basis of their topic. The
respondents were asked to identify the level of their agreements with the statements.
This research aims to draw conclusions about all Russian companies. Since the number of
Russian companies is too large to question all of them, a sample was used as a proxy for the
whole population. Self-selection sampling method was chosen from the variety of sampling
methods to collect cross-sectional data. Self-selection sampling method is a sampling method, in
which individuals take part in the survey voluntarily. Saunders, Thornhill and Lewis (2009) say
that participants often take part in the survey because they are interested in the research topic,
and this phenomenon may positively affect the survey. In case of this research, this feature of
self-selection sampling is indeed beneficial, since in order to answer correctly the questions in
the questionnaire, the respondent should be interested in innovation and, particularly, in the
innovations in his company. Otherwise, he cannot know the data required to answer the
questionnaire.
Self-selection sampling is a non-probability sampling method. This feature makes it
tougher to generalize about the population. However, it is not rational and almost impossible to
use any probability sampling technique in this research. A disadvantage of random sampling
technique in Russia is that Russian companies do not want to spend time participating in
scientific research, especially if research is conducted by a student for his thesis. This situation
leads to very low response rate that makes sample not representative, so this being not
representative eats the benefits of being a probability sample.
The number of observations for regression analysis should be at least the number of
independent variables in the model multiplied by ten – fifteen (Field 2013). Therefore, product
innovation model should include at least 80 observarians, and precess innovation model should
include at least 70 observations. Nethertheless, having anticipated that people would not respond
56
to all required questions and that some answers should be excluded from the regression, the
researcher collected more that 80 responses: the researcher collected 148 responses. After
stepwise exclusion, the product innovation model had 99 observations, and the process
innovation model had 82 observations – just as many as were required Feild (2013).
The research was not designed to capture industry-specific differences in innovation
performance, so for the sake of concision, questions about the industry of the companies were
not asked. The only question that was asked is the question about the company’s size. The data
about the size of the companies in the sample is presented on the Figure 25.
Figure 25. The size of the companies in the sample
The number of men and women was balanced (see Figure 26). However, the sample was
dominated by young (20-23 years old) people (see Figure 27), because the researcher mainly
asked his acquaintances to participate in the survey. Nevertheless, the domination of young
respondents in the sample does not distort the result of the research, since the research is
dedicated to study not people, but companies, while people were only sources of the required
information.
57
Figure 26. Sex of the respondents
Figure 27. Age of the respondents
58
2 RESULTS OF THE REGRESSION ANALYSIS, DISCUSSION
AND CONCLUSION
2.1 Statistical results of the regression analysis
After data collection, two linear regression models were created in IBM SPSS v. 22 –
product innovation model and process innovation model. Table 6 presents summary statistics for
product innovation model. The R2 and adjusted R2 of the model (.413 and .361, respectively) are
indicative of a reasonably well specified model. Accordingly, the F-statistic for the regression
model as a whole is significant (F = 7.992, p = 0.000) at less than the 1 percent level.
Table 6. Cross-sectional regression model of product innovation
Unstandardized
Standardized
Coefficients
Coefficients
Independent
variables
B
Std. Error
Beta
(Constant)
-3.006
9.795
ResandDev
.702***
.180
.335
Patents
-.671
.591
-.100
NewEquip
-.023
.373
-.005
Training
-1.183**
.515
-.204
Marketing
1.761***
.433
.371
StrategyMedian
8.983***
2.409
.394
BIMedian
-.149
2.832
-.005
IdeasMedian
-2.086
2.598
-.087
7.992
F-statistic
2
R
0.413
2
Adjusted R
0.361
a. Dependent Variable: ProdRev
b. ** coefficients significant at the .05 level
*** coefficients significant at the .01 level
t
-.307
3.907
-1.136
-.062
-2.297
4.070
3.728
-.053
-.803
Sig.
.760
.000
.259
.951
.024
.000
.000
.958
.424
0.000
Collinearity
Statistics
Tolerance
VIF
.878
.825
.816
.822
.778
.577
.643
.550
1.139
1.212
1.226
1.216
1.286
1.734
1.555
1.819
Table 7 presents summary statistics for process innovation model. The R2 and adjusted
R2 of the model (.231 and .159, respectively) are indicative of a reasonably well specified model.
Accordingly, the F-statistic for the regression model as a whole is significant (F = 3.222, p =
0.005) at less than the 1 percent level.
59
Table 7. Cross-sectional regression model of process innovation
Unstandardized
Standardized
Coefficients
Coefficients
Independent
variables
B
Std. Error
Beta
(Constant)
4.593
3.067
ResandDev
-.013
.076
-.019
Patents
-.121
.182
-.073
NewEquip
-.034
.105
-.034
Training
.414***
.151
.308
StrategyMedian
2.416***
.707
.424
BIMedian
-.494
.807
-.072
IdeasMedian
-1.228
.781
-.199
3.222
F-statistic
2
R
0.231
2
Adjusted R
0.159
a. Dependent Variable: ProcRev
b. ** coefficients significant at the .05 level
*** coefficients significant at the .01 level
t
1.498
-.177
-.664
-.325
2.738
3.420
-.612
-1.571
Sig.
.138
.860
.509
.746
.008
.001
.542
.120
0.005
Collinearity
Statistics
Tolerance
VIF
.846
.837
.924
.808
.668
.741
.642
1.182
1.195
1.082
1.237
1.496
1.349
1.557
2.2 Discussion of the results
At the beginning of this research, the model of determinants of innovation performance in
Russian companies was created, and eight hypotheses were proposed (see Table 4). At empirical
stage of this research, these hypotheses were tested, using regression analysis. After the
regression analysis, some variables happened to be significant, some not. If the variable was
significant, this fact proved or disproved the initial hypothesis. If the variable was not significant,
this fact did not mean anything, i.e. it neither proved nor disproved the initial hypothesis. Table
8 summarizes the information about the results on hypotheses testing. “+” means that the
determinant is proved to be positively correlated with the type of innovation. “-” means that the
determinant is proved to be negatively correlated with the type of innovation. “Not proven”
means that the regression analysis neither proved nor disproved the hypothesis (the variable was
insignificant). “No hypothesis” means that no hypothesis about correlation between the variable
and the innovation performance was proposed.
60
Table 8. Proved and not proved hypotheses
Hyp.
#
1
2
3
4
5
6
7
8
Process
innovation
Research and experimental development
not proved
Acquisition of other external knowledge
not proved
Acquisition of machinery, equipment and other capital goods
not proved
Training
+
no
Marketing preparation for product innovations
+
hypothesis
Features of the company's innovation strategy
+
+
Features of the company's competitive and market intelligence
not proved
not proved
Features of the company's idea management
not proven
not proved
The regression analysis showed that the more the company invest in research and
Determinants
Product
innovation
+
not proved
not proved
-
experimental development, the higher its product innovation performance is. Indeed, if the
company regularly develops new products, the chances are high that the company comes up with
some innovative product. This phenomenon was noted in many research papers. For instance,
Romijna and Albaladej (2002) proved positive correlation between R&D expenditures and
product innovation success, using even three different measures of product innovation success:
incidence of product innovation, the number of patents held, and product innovation index.
Napolitano (1991) and Leblanc et al. (1997) also emphasize the importance of research and
development for innovation, and the obtained results proved their views.
Beiersdorf, a personal care company that owns such famous brands as Nivea, relies
heavily of internal R&D. Its Hamburg research center employs more than 450 researches, spread
across 16000 square meters and has 150 million euro annual budget. Besides Hamburg center,
the company has around 120 researches around the world (Perepu 2014). Samsung heavily
invests in localized innovation units, called Product Innovation Teams, whose primary
responsibility is to create and implement product innovations (Wedell-Wedellsborg and Miller
2014).
The regression analysis identified negative correlation between training expenditures and
products innovation performance, and positive correlation between training expenditures and
process innovation performance. This fact corroborates the ideas of Kim (1980), Lee, Bae and
Choi (1988). The investments in training make employees’ behavior more effective in terms of
61
process, but less effective in terms of creativity, enhancing process innovation performance and
hampering product innovation performance. Baldwin and Johnson (1995) point out the
importance of training. However, the author claims that training programs are not equally useful
for different companies: they are more efficient for the companies with expertise in innovation
and quality management and less effective for others.
On the other hand, practical examples show the training may negatively affect product
innovation. In 2006, Beiersdorf launched a global consumer connectivity training program called
InTouch, in which the employees were taught to understand the customers better. The company
used case studies, discussion and practical examples to identify consumer needs that can
potentially be incorporated into new products. The training resulted in multitude of product
initiatives launched in more than 50 countries (Perepu 2014). Thus, training of market
intelligence may positively affect product innovation performance.
The regression analysis showed that the more the company invest in marketing
preparation for product innovations, the higher its product innovation performance is. That
conclusion is quite intuitive, since thorough marketing preparation of innovative products
facilitates the product’s adoption by population: it influence knowledge and persuasion steps of
innovation-decision process (Rogers 1962).
Marketing plays especially important role than the new product is controversial and
induces debates over its usefulness and consequences. Giesler (2013) in his article “How
Doppelgänger Brand Images Influence the Market Creation Process: Longitudinal Insights from
the Rise of Botox Cosmetic” studies the history of Botox Cosmetics, a remedy against wrinkles.
The article explores how contestation between brand adepts and opponents can contribute to
brand image and offers a four-step approach to revitalize the brand, using this contestation. The
author studies the history of Botox Cosmetics brand, outlining five periods of it. Each period is
characterized by unique brand positioning, sometimes mutually exclusive, ranging from
“pleasurable play” to “a weapon of liberation”.
In order to reinforce the controversial brand positive image, the company should follow
four steps: problematization, interessment, enrollment, and mobilization. At problematization
step, the company should screen how the brand “restore the harmony” if people use it. For
instance, using Botox may underscore the women’s belief in science and technology, whereas
62
not using Botox may classify her as being outdated and imbued with prejudices. At the next step,
interessmant, the company should enlist the support of experts, such as scientists, doctors and
others that will validate the company’s propaganda. Then, at enrollment step, the company has to
demonstrate the value of the good through concrete customer performance, asking housewives
and mothers to speculate about benefits of using the good. At the last stage, mobilization, the
company should make its current customers to adopt the current brand image and loose the
previous one. To sum up, marketing is important for innovative products, especially if they are
controversial.
The regression analysis showed positive correlation between features of the company's
innovation strategy and product and process innovation performance. This result supports the
findings of previous research. Zien and Buckler (1997) hold that employees and organizational
content are the main drivers of innovation performance. Lee and Na (1994) argue that
management support and commitment for innovation is crucial to innovation performance,
especially in case of radical innovation that may be risky and costly. Without innovation
strategy, communicated to and understood by employees, it is hard to induce such radical
innovations.
The efficient innovation strategy is especially important when being innovative is
harmful for the company in some way. Kodak is a great illustration of this idea: Kodak failed to
develop digital photography technology because Kodak thought that digital photography would
eat current company’s revenues. This situation is called “The Innovator's Dilemma” (Christensen
1997). “The innovators dilemma” is a situation in which a company rejects innovations because
today’s customers cannot use them. The companies are adhered to customer current needs and
disregard innovative ideas that resulting in losing market dominance when customers adopt the
innovation that the “successful” companies have let to go. Christensen argues that investing in
disruptive technologies is not a rational financial decision for senior managers to make because,
for the most part, disruptive technologies are initially of interest to the least profitable customers
in a market. Even though Kodak had great developments in digital photography, it failed to profit
from them because managers were afraid that digital photography would eat current company’s
profits that were great. That is why Kodak failed. Kodak could set up new enterprises to deal
with digital photography and harness its technical knowledge to excel on the new digital
63
photography market, rather than to let the others do so. In order to avoid such failures,
management and employees should follow established and properly delivered innovation
strategy.
The correlation between the features of the company’s market or business intelligence
and innovation performance was proven neither for product nor for process innovation, despite
many researchers suggest that this correlation should exist. As it was mentioned by AtuaheneGima (1996), the relationship between market orientation and innovation has been debated for
decades. Some scholars argue that market orientation negatively influence product innovation
since market orientation leads to the development of the products similar to the competitors’, so
called "me-too" products, rather than real innovations (Bennett and Cooper, 1981). On the other
hand, other scholars hold that market orientation positively affect the company’s innovation
performance (Deshpande et al., 1993; Kohli and Jaworski, 1990; Webster, 1988). However, as in
this study, the researchers could not prove this claim empirically (Lawton and Parasuraman,
1980).
Atuahene-Gima (1996) found that the company’s market orientation negatively affects
product newness because market orientation prevents radical innovations (see Figure 28).
Product newness, in turn, is negatively correlated with market success. On the other hand, the
company’s market orientation positively affects product advantage, innovation-marketing fit and
interfunctional teamwork, all of which are positively correlated with marketing success. To sum
up, market orientation makes products less innovative, but more successful on the market.
64
Figure 28. A model by Atuahene-Gima (1996). The numbers on the figure are standardized beta
weights.
The correlation between the features of the company’s idea management and innovation
performance was proven neither for product nor for process innovation, despite almost all
researches suggest that this correlation should exist. For instance, Nonaka and Takeuchi (1995)
argue that the information sharing stimulates creativity in the organization, and creativity, in
turn, helps the company to create more innovative products. Therefore, the companies should
motivate their employees to share their ideas.
Prajogo and Ahmed (2006) say that innovation performance requires the context: the
company should actively promote the idea of innovation and create idea-friendly environment,
that is the companies should “develop managerial practices and actions that function as a
65
stimulus for encouraging and energizing people to innovate through development and
accumulation of ideas and knowledge”.
Furthermore, the companies should also pay attention to external ideas, taking into
account the concept of open innovation (Chesbrough 2003). Laursen and Salter (2006)
conducted a vast survey that included a sample of 2707 manufacturing firms in the UK. They
found out that the companies that exhibits width and depth in their external idea search tend to be
more innovative. However, the correlation is U-shaped (see Figure 29).
Figure 29. Predicted relationship between innovative performance and the breadth of search
through external sources. Source: Laursen and Salter (2006)
2.3 Theoretical contribution, limitations and further research
During this research, the attributive model of innovation at the firm’s level was created.
This model addresses the following issues of the previous models:
First, previous models provide one best way of innovation process, eliminating
alternative paths (2002, cited in Hobday (2005)). Mahdi argues that evidence demonstrates that
66
major differences in innovation process exist within and across different industries. Moreover,
these differences persist over time and are not a deviation from a norm. My model does not
provide one best way of innovation process, eliminating alternative paths, but rather enlist
drivers that make the company successful in innovation. In AMI model in does not matter
whether, for instance, market need or technological development comes first, since its nonsequential nature.
Second, most of the models assume that the companies behave too rationally, being able
to hypothesize a solution to an innovation problem, such as a new product development, and then
systematically solve the problem, using a standard toolkit such as design thinking, prototype
testing and market research. However, this case is only possible when companies have enough
experience to make an educated guess about the potential solution, otherwise, they have to use
iterative approach, experimenting, making mistakes, and trying again. Non-sequential nature of
the attributive model eliminates issues with hype-rationality, because non-sequential nature
allows for trial-and-error experimentation and rudimentary activities.
Third, the models lack a coherent theoretical base, which is important because it can help
to put innovation within the wider organizational and strategic context in which it belongs.
Indeed, the state-of-the-art models treat innovation is separate process, failing to understand that
it is tightly connected with the firm’s strategy, culture and capabilities. The attributive model
rests on a solid theory, namely modern resource-based theory.
Forth, all the models deal with innovation leaders, neglecting latecomers. This fact is
especially important for scientist who research innovation in developing countries, e.g. Russia,
since many companies there do not develop innovations themselves, but rather adopt them from
abroad. Moreover, the majority of models deals describe processes in the large corporations, not
paying attention to medium and small companies, where innovation process usually do not have
any formal stages and domain. For instance, a small company may not even have R&D
department. Therefore, for such companies it is definitely inappropriate to apply existing models.
The attributive model fits both innovation leaders and latecomers, and both large corporations
and small companies, because the determinants of AMI neglect organizational structure and
encompass sources of innovations far beyond only R&D department.
67
The determinants of innovation performance were tested on the cross-sectional sample of
148 Russian companies. The correlation between some variables and innovation performance
was proved statistically. That means that the attributive model is grounded not only on the
existing theory, but also on the empirical data. However, in this study, causation was not proved,
i.e. it was not proved that the determinants elicit innovation performance, not vice versa. The
causation proof may be interesting for further studies.
The current analysis was a high-level, not industry specific analysis. However, the
differences in innovation processes across the industries do exist, and it may be interesting to
adjust the attributive model for different industries.
2.4 Practical contribution
Based on the research, the following recommendation can be given to the management of
companies that would like to boost innovation performance. This study produced the model of
innovation performance at the firm level. Despite causation between the determinants and the
company’s innovation performance has not been proved and in regression analysis, some
variables occurred to be insignificant, the attributive model can make it more clear for managers
which factors are connected with the company’s innovation performance. The managers may use
the attributive model as a starting point if they want to make their companies more innovative.
They may use logical reasoning to understand which determinants may be more influential for
their industry and their company, identify the weak places in their company (e. g. use of external
ideas in the innovation process), and craft a plan to overcome them. To sum up, the attributive
model is tool that, after some additional work, managers can use to make their company more
innovative.
68
3 REFERENCES
Acs, Zoltan J, and David B Audretsch. 1988. "Innovation in Large and Small Firms: An
Empirical Analysis." The American Economic Review 678-690.
Albaladejo, Manuel, and Henny Romijn. 2002. "Determinants of innovation capability in small
electronics and software firms in southeast England." Research Policy 1053–1067.
Atuahene-Gima, Kwaku. 1996. "Market Orientation and Innovation ." Journal of Business
Researc 93-103.
Bain
and
Company.
2011.
"The
great
eight."
Bain
&
Company
web
site.
http://www.bain.com/Images/BAIN_BRIEF_8MacroTrends.pdf.
Baldwin, John R., and Joanne Johnson. 1995. Human Capital Development and Innovation: The
Case of Training in Small and Medium-Sized Firms. Statistics Canada Working Paper.
Bass, Frank M. 1969. "A new product growth for model consumer durables." Management
Science 215-227.
Bearda, Charles, and Chris Easingwood. 1996. "New product launch: Marketing action and
launch tactics for high-technology products." Industrial Marketing Management 87–103.
Bonnin, Gaël, Olivier Segard, and Pierre Vialle. 2005. "Relationship Marketing and Innovation:
The Case of the Launch of Wireless Local Loop Telecommunication Services in France."
Journal of Services Research 149-171.
Cassiman, Bruno, and Veugelers Reinhilde. 2002. "Complementarity in the Innovation Strategy:
Internal R&D, External Technology Acquisition and Cooperation."
Chandrasekaran, Deepa, and Gerard J. Tellis. 2011. "Getting a Grip on the Saddle: Chasms or
Cycles?" Journal of Marketing 21-34.
Chesbrough, Henry W. 2003. Open innovation: The new imperative for creating and profiting
from technology.
Christensen, Clayton. 1997. The Innovator's Dilemma. Harvard Business Review Press.
69
Cooper, Robert G. 2008. "The Stage-Gate. Idea-to-Launch Process–Update, What's New and
NexGen Systems." Journal of Product Innovation Management 213-232.
Cooper, Robert G., and Scott J. Edgett. 2012. "Benchmarking Best Practices: Performance
Results and the Role of Senior Management."
Crepon, Bruno, Emmanuel Duguetb, and Jacques Mairessec. 2006. "Research, Innovation And
Productivity: An Econometric Analysis At The Firm Level." Economics of Innovation
and New Technology 115-158.
Cumming, B.S. 1998. European Journal of Innovation Management.
Davis, Fred D. 1989. "Perceived usefulness, perceived ease of use, and user acceptance of
information technology." MIS Quarterly 319–340.
Drucker, Peter F. 1985. "The discipline of innovation." Harvard Business Review 67-72.
Emmons, Garry, Julia Hanna, and Roger Thompson. 2012. Five Ways to Make Your Company
More Innovative.
Entrepreneur.
n.d.
Competitive
Intelligence.
Accessed
September
8,
2015.
http://www.entrepreneur.com/encyclopedia/competitive-intelligence.
Field, Andy. 2013. Discovering Statistics using IBM SPSS Statistics .
Floorwalker, Mike. 2013. 10 Innovative Pieces of Technology That Failed Miserably. January
12. Accessed September 27, 2015. http://listverse.com/2013/01/12/10-innovative-piecesof-technology-that-failed-miserably/.
Freeman, C. 1987. "Technology Policy and Economic Performance." 129. London: Pinter.
Giesler, Markus. 2013. "How Doppelgänger Brand Images Influence the Market Creation
Process: Longitudinal Insights from the Rise of Botox Cosmetic." Journal of Marketing.
Gilad, Benjamin. 2003. Early Warning: Using Competitive Intelligence to Anticipate Market
Shifts, Control Risk, and Create Powerful Strategies.
Gupta, Praveen K., and Rajesh K. Tyagi. 2008. A Complete and Balanced Service Scorecard:
Creating Value Through Sustained Performance Improvement.
70
Hippel, Eric. 2006. Democratizing Innovation. MIT Press.
Hobday, Michael. 2005. "Firm-level Innovation Models: Perspectives on Research in Developed
and Developing Countries." Technology Analysis & Strategic Management 121–146.
Kim, Linsu. 1980. "Stages of development of industrial technology in a less developed country: a
model." Research Polic 254–277.
Kishore, Sandeep. n.d. "The power of incremental innovation." Wired. Accessed September 22,
2015. http://www.wired.com/insights/2013/11/the-power-of-incremental-innovation/.
Kommersant. 2016. "Инновации бьют рублем." March 22.
Laursen, Keld, and Ammon Salter. 2006. "Open for innovation: the role of openness in
explaining innovation performance among U.K. manufacturing firms." Strategic
Management Journal 131-150.
Lee, J, Z Bae, and D Choi. 1988. "Technology development processes: a model for a developing
country with a global perspective." Research and Development Management 235–250.
Lewis, Joanna I. 2007. "Technology Acquisition and Innovation in the Developing World: Wind
Turbine Development in China and India." Studies in Comparative International
Development 208-232.
Mahdi, S. 2002. "Search strategy in product innovation process: theory and evidence from the
evolution of agrochemical lead discovery process." University of Sussex.
Manheim, Jarol B, and Richard C. Rich. 1995. Empirical Political Analysis: Research Methods
in Political Science. Longman.
Maryanenko, Viktor P. 2008. Феномен инновации. Санкт - Петербург.
Maurer, Indre. 2010. "How to build trust in inter-organizational projects: The impact of project
staffing and project rewards on the formation of trust, knowledge acquisition and product
innovation." International Journal of Project Management 629–637.
Moore, Geoffrey A. 1991. Crossing the Chasm: Marketing and Selling High-Tech Products to
Mainstream Customers. Harper Business Essentials.
71
OECD. 1997. Oslo Manual, 2nd edition. Oslo.
OECD. 2005. Oslo Manual, 3rd edition. Oslo.
Paap, Jay, and Ralph Katz. 2004. "Anticipating disruptive innovation." Research Technology
Management (Research Technology Management).
Perepu, Indu. 2014. "Open Innovation at Beiersdorf – The Launch of Nivea Invisible for Black
& White." IBS Center for Management Research.
Prajogo, Daniel I, and Pervaiz K Ahmed. 2006. "Relationships between innovation stimulus,
innovation capacity, and innovation performance." R&D Management 499-515.
PwC. 2013. "Global Innovation Survey: Innovation, growth and business strategy."
Robinson, Les. 2009. A summary of diffusion of innovations. January. Accessed January 2009.
http://www.enablingchange.com.au/Summary_Diffusion_Theory.pdf.
Rogers, Everett. 1962. Diffusion of Innovations.
Romijna, Henny, and Manuel Albaladej. 2002. "Determinants of innovation capability in small
electronics and software firms in southeast England." Research Policy 1053–1067.
Rothwell, Roy. 1994. "Towards the Fifth‐generation Innovation Process." International
Marketing Review 7-31.
Russian Government official web-site. 2016. Innovations. Accessed March 25, 2016.
http://government.ru/govworks/28/main/.
Rycroft, Robert W, and Don E Kash. 2000. "Steering Complex Innovation." Research
Technology Management 18-23.
Saunders, Mark, Adrian Thornhill, and Philip Lewis. 2009. Research methods for business
students. Harlow: Pearson Education Limited.
Schumpeter, Joseph. 1942. Capitalism, Socialism and Democracy. New York: Harper.
Teece, David J., Gary Pisano, and Amy Shuen. 1997. "Dynamic Capabilities and Strategic
Management." Strategic Management Journal 509-533.
72
Utterback, James M, and William J Abernathy. 1975. "A Dynamic Model of Process and Product
Innovation." The International Journal of Management Science.
Van den Bulte, Christophe. 2005. "Want to know how diffusion speed varies across countries
and products? Try using a Bass model." Visiona magazine.
Wedell-Wedellsborg, Thomas, and Paddy Miller. 2014. "Samsung's European Innovation Team."
IESE, March 17.
Whipps, Heather. 2008. How the Steam Engine Changed the World. June 16. Accessed
September
26,
2015.
http://www.livescience.com/2612-steam-engine-changed-
world.html.
Yli-Renko, Helena, Erkko Autio, and Harry J Sapienza. 2001. "Social capital, knowledge
acquisition, and knowledge exploitation in young technology-based firms." Strategic
Management Journal 587–613.
73
4 APPENDICES
Appendix 1
Figure 30. Examples of the general questions
74
Appendix 2
Figure 31. An example of the question about the company's financial ratios
75
Appendix 3
Figure 32. An example of the questions about the company’s processes
76
Отзывы:
Авторизуйтесь, чтобы оставить отзыв