Demand and Design Choices in an Open Innovation system: The case for CoPS and B2B Virginia Acha CoPS Centre, SPRU & CENTRIM (Us of Sussex &Brighton) Presentation.

Demand and Design Demand and Design Choices in an Open Choices in an Open Innovation system:Innovation system:

The case for CoPS and The case for CoPS and B2BB2BVirginia AchaCoPS Centre, SPRU & CENTRIM (Us of Sussex &Brighton)

Presentation to the CIS User GroupPresentation to the CIS User GroupDTI Innovation Economics ConferenceDTI Innovation Economics ConferenceNovember 17, 2006November 17, 2006

Work in ProgressWork in Progress

Project Team

Centre for Complex Products and Systems (CoPS)

- Virginia Acha- Mike Hobday- Howard Rush

CENTRIM, University of Brighton

Overview

Demand and Design Choices in an Open Innovation System Research Aims Project Methodology

Descriptive findings - B2B and CoPS Characteristics in the UK population Innovative profile

Open Innovation in CIS4 Through the B2B and CoPS lenses

Drivers for Open Innovation patterns Role of design and market dynamics Models and preliminary results

Limitations and Conclusions

Demand and Design Choices in an Open Innovation system Research Aims

Open innovation models used to describe increasingly complex and distributed patterns of innovation (Chesbrough, 2003; von Hippel and von Krogh, 2003; Coombs, Harvey and Tether, 2001)

Pattern that has been core to the development of Complex Products and Systems (CoPS)

o Emergence of systems integration and integrated solutions in response

o Core role of design and customer engagement in these responses

Complex Products and Systems (CoPS) High value, engineering-intensive customised capital goods Produced in one-off projects or small batches (Hobday, 1998) Decade of empirical, largely case study research (ESRC Centre,

www.cops.ac.uk) Economic impact and classification (Acha et al, 2004)

Extend to B2B CoPS as a leading sector of B2B innovation Can open innovation patterns be found in both

o CoPSo B2B

Drivers for Open Innovation Characteristics of CoPS innovation and production patterns

lend themselves to open structures Design, customer engagement, undefined markets as drivers

Test these questions using the CIS4 Define B2B (Q3) Define CoPS

o Classification system (Acha et al, 2004)o CoPS-based Services classification also done in previous study,

now applicable

Demand and Design Choices in an Open Innovation system

Overview

Demand and Design Choices in an Open Innovation System Research Aims Project Methodology

Descriptive findings - B2B and CoPS Characteristics in the UK population Innovative profile

Open Innovation in CIS4 Through the B2B and CoPS lenses

Drivers for Open Innovation patterns Role of design and market dynamics Models and preliminary results

Limitations and Conclusions

11721

4723

Key Descriptives: B2B

Firms by main customers - B2B* 71% of all respondents

B2B or B2G

B2C188

152

60

BothB2B & B2G

Both B2B & B2C All

Question 3 is open to some interpretation, as some firms

recorded mixed markets.

Key Descriptives: CoPS are B2B

By definition, CoPS are B2B, aren’t they? Based on established classification (Industrial and Corporate Change,

2004)

Addition of CoPS-based services CoPS classified respondents cross-checked by main

customers 356 firms identified as CoPS but with B2C markets, of which

o 1717 also have B2B customers, ando 1111 have B2G customers

Seven SIC codes account for 78% of this crossovero In services and constructiono 3 dropped from the CoPS filter - ambiguouso 4 retained: queried share small in comparison, clear CoPS

relevance (e.g. telecommunications, engineering consultancy)

14100

986

1359

2345

Key Descriptives: CoPS

Manufacturing & Construction

ServicesCoPS

CoPS firms represent 14% of the survey population.

Patterns in Innovation

0.13

0.22

0.27

0.30

0.25

0.14

0.20

0.29

0.18

0.37

0.20

0.33

0.40

0.36

0.43

0.14

0.23

0.25

0.22

0.28

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

B2C B2B CoPS all CoPS Production CoPS Services

Shar

e of

Pop

ulat

ion

Goods Innov Services Innov Product Innov Process Innovator

B2B - greater innovators

CoPS even more so

Product, Process Innovation in CoPS Services-Integrated Solutions

CoPS Services leading in service innovations

0.53

0.60

0.660.68

0.65

0.90

0.850.82

0.78

0.85

0.26

0.310.34 0.33 0.34

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

B2C B2B CoPS all CoPS Production CoPS Services

Sh

are

of

Pro

du

ct o

r P

roce

ss I

nn

ovato

rs

Product New to Market Product New to Enterprise Process New to Industry

Degree of novelty in Innovation

Greater novelty in B2B and CoPS

Overview

Demand and Design Choices in an Open Innovation System Research Aims Project Methodology

Descriptive findings - B2B and CoPS Characteristics in the UK population Innovative profile

Open Innovation in CIS4 Through the B2B and CoPS lenses

Drivers for Open Innovation patterns Role of design and market dynamics Models and preliminary results

Limitations and Conclusions

Relative openness in Product Innovation

B2C relatively more open to external sources

CoPS more collaborative

But 2/3rds still mainly done in-house

0.61

0.69 0.69 0.69 0.69

0.22 0.22 0.24 0.24 0.23

0.16

0.08 0.07 0.07 0.08

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

B2C B2B CoPS all CoPSProduction

CoPSServices

Sh

are

of

Pro

du

ct I

nn

ovato

rs

Internal Development (Share of Total)Collaborative Development (Share of Total)External Development (Share of Total)

0.52

0.60

0.66

0.61

0.68

0.34

0.280.26

0.33

0.22

0.140.12

0.08

0.05

0.10

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

B2C B2B CoPS all CoPS Production CoPS Services

Sh

are

of

Pro

cess I

nn

ovato

rs

Internal Development Collaborative Development External Development

Relative openness in Process Innovation

AgainB2C relatively more open to external sources

CoPS do more in-house

Openness in Innovation Activities

0.23

0.35

0.43

0.39

0.45

0.10

0.13

0.17 0.170.17

0.12

0.15

0.19

0.16

0.21

0.14

0.20

0.26

0.29

0.23

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

B2C B2B CoPS all CoPS Production CoPS Services

Sh

are

of

po

pu

lati

on

Intramural R&D Extramural R&D Acquisition of external knowledge Design

** significance, except for CoPS production, Acquisition of External Knowledge

B2B more open in unlinked innovation activities

CoPS more open and do more in-house

Design features prominently

Breadth and Depth of Information Sources

4.27

5.86

6.39

6.04

6.65

2.90

3.57

4.13

3.78

4.38

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

B2C B2B CoPS all CoPS Production CoPS Services

Avera

ge n

um

ber

of

so

urc

es

All sources Breadth & Depth

** significance for all Laursen & Salter (2006)

All sources - count

Breadth & Depth - count where Medium or High

B2B and CoPS more intensive users of sources of information

Overview

Demand and Design Choices in an Open Innovation System Research Aims Project Methodology

Descriptive findings - B2B and CoPS Characteristics in the UK population Innovative profile

Open Innovation in CIS4 Through the B2B and CoPS lenses

Drivers for Open Innovation patterns Role of design and market dynamics Models and preliminary results

Limitations and Conclusions

What leads to Open Innovation? Descriptive evidence shows B2B and CoPS firms as

relatively open innovation systems Collaboration Drawing externally for innovation

o More for CoPS than B2B for innovationso Both in innovation activities

More intensive users of informationo Breadth and Depth (Laursen & Salter, 2006)

What drives the process to open innovation? Chesbrough (2003) reflects upon

o Markets for ideas and technologyo Availability and mobility of human capitalo Distributed and enhanced capability across the value chain

What leads to Open Innovation? Tendency to open innovation patterns may also be related to:

Nature of innovation in the firm Market dynamics for the firm

Nature of innovation in the firm Partitioning of the innovation process Design as a translator, bridge across stages, sectors,

specialisationso Role of design (Tether, DTI Presentation, 2006; Whyte, Bessant

and Neely, 2005)o In partitioning (von Hippel, 1990)

Market dynamics for the firm Uncertainty, unknown markets Engagement with the consumer

Testing for Open Characteristics Openness

How to construct a variable? Characteristics of openness across questionnaire

o How innovations are developed (Q6, Q10) Innovation through collaboration Innovation through others

o Innovation activities (Q13) and values (Q14) Acquisition of R&D or Acquisition of External Knowledge

o Sources of information (Q16) Breadth and depth

o Co-operation (Q17, Q18) Correlations show interesting distinctions

Open Activities - selected as a proxy

Correlations

Open activities

Open Sources Collaborate

Innovation through

Collaboratio

Innovation through others

Open activities

Pearson Correlation 1.00 0.36 0.30 0.25 0.07Sig. (2-tailed) . 0.00 0.00 0.00 0.00N 16445.00 15698.00 15753.00 16445.00 16445.00

Open Sources

Pearson Correlation 0.36 1.00 0.29 0.21 0.08Sig. (2-tailed) 0.00. 0.00 0.00 0.00N 15698.00 15698.00 15692.00 15698.00 15698.00

CollaboratePearson Correlation 0.30 0.29 1.00 0.27 0.06Sig. (2-tailed) 0.00 0.00. 0.00 0.00N 15753.00 15692.00 15753.00 15753.00 15753.00

Innovation through

Collaboration

Pearson Correlation 0.25 0.21 0.27 1.00 -0.01Sig. (2-tailed) 0.00 0.00 0.00. 0.14N 16445.00 15698.00 15753.00 16445.00 16445.00

Innovation through

othersPearson Correlation 0.07 0.08 0.06 -0.01 1.00Sig. (2-tailed) 0.00 0.00 0.00 0.14.N 16445.00 15698.00 15753.00 16445.00 16445.00

**Correlation is significant at the 0.01 level (2-tailed).

Measures of Openness

Testing the sources: Design

Openness = ƒ(importance of design) Proxies

Design activities (Q13) Registration of design (Q21) Complexity of design (Q21)

Predicted positive relationships with open activities

Logistic regression 15,699 used in analysis Size (employment), B2B, CoPS

Design model results

Proxies are positively related, as predicted. Design activities and Design complexity influential and positively

correlated with open activities. Registration less influential

CoPS and size weakly positive; B2B weakly negative

B SE Sig. Exp(B)Design activities 1.511 0.049 0.000 4.53Design Registration 0.000

Low 0.188 0.075 0.012 1.207Medium 0.289 0.08 0.000 1.335

High 0.382 0.086 0.000 1.466Complexity of Design 0.000

Low 0.84 0.065 0.000 2.315Medium 0.891 0.069 0.000 2.439

High 1.113 0.089 0.000 3.042Size (SME, Large) 0.299 0.051 0.000 1.349B2B -0.074 0.052 0.155 0.929CoPS all 0.168 0.06 0.005 1.182Constant -2.168 0.047 0.000 0.114Note: Pseudo R2=.23 (Nagelkerke). Model Chi-square (1)= 2531, p<.000

Testing the sources: Market Dynamics Openness = ƒ(Importance of customer

engagement, ‘cloudy’ markets) Proxies

Clients as a source of information (Q16) Clients as collaborators (Q18) Lack of information on markets (Q19) Uncertain demand (Q19)

Predicted positive relationships with open activities

Logistic regression 15,684 used in analysis Size (employment), B2B, CoPS

Market Dynamics model results

Proxies are positively related, as predicted.

Client engagement positive and influential

‘Cloudy markets’ positive but only weakly influential

CoPS and size weakly positive; B2B weakly negative

B SE Sig. Exp(B)Clients as a Source 0.000

Low 1.337 0.095 0.000 3.809Medium 1.524 0.086 0.000 4.59

High 1.714 0.085 0.000 5.552Client Collaborators 1.085 0.056 0.000 2.958Uncertain Demand

Low 0.45 0.066 0.000 1.569Medium 0.468 0.069 0.000 1.596

High 0.365 0.088 0.000 1.441Lack of market info

Low 0.32 0.061 0.000 1.377Medium 0.58 0.071 0.000 1.786

High 0.451 0.123 0.000 1.571Size (SME, Large) 0.409 0.049 0.000 1.505B2B -0.059 0.051 0.251 0.943CoPS all 0.244 0.058 0.000 1.276Constant -3.393 0.081 0.000 0.034Note: Pseudo R2=.227 (Nagelkerke). Model Chi-square (1)= 2469, p<.000

Limitations and ‘To Do’

Fine tune the individual models Some noise in the results Comparison of B2B, CoPS, CoPS production, CoPS services

sub-groups Link the models

Only a partial answer to the question, ‘What drives open innovation processes?’ Design, Market dynamics +… Proxies are limited

o Time dimension to demonstrate dynamics Causality cannot be proven

Only correlation Other research methods needed to establish direction of link

Conclusions

CoPS are relatively more open in innovation B2B show more open innovation practices (activities,

practices) B2C innovations acquired more externally Measures of ‘openness’ show interesting distinctions

o Choice is meaningful

Features that have contributed to open structures in CoPS may be drivers in all sectors

Evidence from design and market models Needs confirmation of a linked, more fully specified model Needs further qualitative search to establish process,

causality

Conclusions

Policy implications Systems Integrators, design play important

connecting roles in an ‘open innovation’ environment Drivers for openness beyond opportunity

o Beyond availability of resources and markets for knowledge

o Perhaps more structural features which will shape the degree to which innovation systems become open

Role for policy

Questions and Discussion