Supplementary files are available for this work. For more information about accessing
these files, follow the link from the Table of Contents to "Reading the Supplementary Files”.
ble of Contents
Business Transformation Success Dynamics; It’s the
Communication
Mark Heffernan, Phillip Wing, G eoff McDonnell
International System Dynamics Pty Ltd, Technology Venture Partners Pty Ltd,
Adaptive Care Systems Pty Ltd
382 Bronte Rd Bronte NSW 2024 Australia
Phone +612 9386 0993 Fax +612 9386 0992
Email: gmcdonne@ bigpond.net.au
ABSTRACT: A Study of Business Transformation(BT) Success involving fifteen
corporations framed business transformation as a type of large-scale strategic
renewal implemented via a diffusion of innovation process within a socially
constructed organisational environment. Based on the qualitative and quantitative
analysis of the communication patterns examined in this study, the complex
relationships were abstracted and reflected in an adapted diffusion of innovation
dynamic model that has communication effectiveness as a key influencing variable.
The key findings of the communications network analysis(CNA) showed that success
was correlated with average path length, opinion leadership and the presence of weak
ties, with a saturation effect on communication after a certain network density is
reached (referred to in CNA literature as percolation effect).
Simulation runs replicating observed BT Success Index measures with group size,
activity, average path length, connectedness and leadership, adjusted for quit
percentage per year showed that with one variable missing the successful
implementation is delayed, two variables missing results in failure and that a
saturation point for communication exists.
KEYWORDS: INNOVATION DIFFUSION, BUSINESS TRANSFORMATION,
SOCAL NETWORK ANALYSIS, COMMUNICATIONS
Introduction
Business Transformation (BT) is an type of large scale organisational change which
results in sustained improvement in capability to meet changing customer or market
needs (Hammer and Champy, 1993).. This form of strategic renewal is usually
enabled by information technology, which allows design of business processes and
requires skilful management of the organization and its people to be successful
(Davenport 1993, O’Neill and Sohal, 1999).
The science of complexity has a number of insights to offer about the nature of
interactions in human systems (Waldrop 1993, McElroy 2000). A subset of
complexity science is complex adaptive systems (CAS) theory, which holds that
living systems, including organizations, self organise and continuously fit themselves
to ever changing conditions in the environment. CAS theory states that this occurs by
modifying their knowledge of fact and practice as a consequence of their interaction
with their environment and the feedback effects of their own and others interaction
(Holland 1995). The traditional, deterministic view of change, is where people as
collections of objects can be manipulated to form social systems that create new ideas
and organisational capabilities. The complex, dynamic view starts with knowledge in
organizations being an emergent social process. Human social systems give rise to
collective knowledge making by their members as a by-product of their
communication, learning and interaction. CAS are said to be driven by three control
parameters, the rate of information flow through the system, the richness of the
connectivity between agents in the system and the level of diversity of the agents
(Stacey 1996).
This view of change is consistent with the diffusion of innovation literature (Rogers
1995), which has been represented in system dynamics world as a form of epidemic
spread, such as the Bass diffusion model (Bass 1969, Sterman 2000).
Communications has always featured strongly in these models, and more structured
techniques of communication network analysis have explored the structure, rate and
quality of communication, within and among the groups comprising organizations
(Scott 1991, Nohria and Eccles 1992, Scott 1996, Krebs 2000). This paper describes
an adaptation of the diffusion of innovation model to incorporate results of a study,
performed by one of the authors (Wing 2001), comparing the success of large BT
projects in 15 companies with the structure of their communications.
Study Overview
An international study of 15 companies undertaking BT projects was designed to
investigate the variables which explain BT success. A BT success index (BTSI) was
constructed based on the implementation time and level of capability achieved,
adjusting for the complexity of the project (Wing 2001). Data was collected using
structured questionnaires administered to project team members and independent
experts. Quantitative and qualitative communication network analysis was performed
on the BT project team using Inflow software (Krebs,1995). These results were
correlated with the BTSI and showed communication as the pivotal explanatory
variable in explaining BT success. The elicited communications pattems were
incorporated into previous system dynamics modelling work jointly performed by the
authors on diffusion of innovation, using ithink software.
Communication Network Analysis
This consisted of calculation of measures of centrality and connectedness, including
group size, average path length, degree, closeness, reach, and activity, and the
visualisation of communication structures, via the use of network maps or sociograms
(Krebs 1996). The network maps are presented based on the following conventions:
The sociogram represents the entire population of people involved in the BT project;
Each block represents an organisational sub-group (i.e. department or workgroup);
The name associated with each organisational sub-group is the companies
abbreviation for that group, e.g. ‘executives’ is the executive team, ‘IT’ is the IT
department;
‘Exterals’ are those groups of people that the BT team communicate with that are
outside the organization, e.g. academics, consultants;
Full lines are two-way confirmed communication;
Dotted lines are one-way communication.
The three figures that follow this example depict the network maps for the lowest,
highest and mid-range BTSI companies from this study.
Figure 1: Lowest BTS! - Company 76
Company 76 - Question 6 All Ties
Figure 2 : Mid Score BTSI| - Company 62
‘Company 62.- Question 6 All Ties
aun aun, sast.50
Figure 3 : Highest BTS| - Company 53
Company 53 - Question 6 All Ties
CNA Results
1,Presence of Weak Ties
Lower levels of network hierarchy (as defined by the predominance of hub and spoke,
or interlocking personal networks, structures) are associated with higher levels of BT
success. This is an indicator that the presence of weak ties within the network is
associated with higher levels of BT success.
2. Locus of Communication
Lower levels of network fragmentation (as defined by the isolation of individuals or
groups from the rest of the network) are associated with higher levels of BT success.
This also supports the existence of weak ties being associated with higher levels of
BT success, as weak ties add a bridge between homophilous or fragmented groups.
3. Opinion Leadership within the Network
Opinion leadership (as defined by the communication activity around an individual) is
associated with higher levels of BT success.
4. Network Density - ‘The Percolation Effect’
Network density (as defined by the volume of communication flows across the
network) is associated with higher levels of BT success up to a threshold level, at
which point, increased levels of density are associated with decreasing levels of BT
success.
System Dynamics Model of BT Success
Some of the specific relationships embodied in the model are:
1. Adopters
* Three classes - potential, active and former
+ Potential adopters interact with former adopters of innovation through weak
ties.
2. Innovation
* abase level of innovation attractiveness exists
¢ this is modified by the leadership within the system.
3: Percolation Effect
The study reported potential diminishing retums on the level of communication
activity. This could be due to either information overload or too much
communicating, not enough doing. This dynamic is also captured in the systems
model by identifying optimum activity and the feedback on “infectivity” or influence
rates
4. The innovation must not only be implemented and used it must continue to
deliver the planned business benefits. The model allows for a ongoing weak tie
connection and a honeymoon period dynamic. This attempts to cater for the ongoing
learning process embodied in the structure and dynamics of the model.
The system dynamics model has been developed as an eight sector model . The
following table provides an overview of the relationships and variables embodied in
each of these sectors of the model. A graphical representation of the model , as
produced by “ithink”, follows this table.
Table 8.6.1
Model Sector
Variables and Relationships
Source
1. Adoption of
= 3 agents- potential, active and
= Bass model
Innovation former adopters. of diffusion
= All agents have quit rates from the |= Infection
innovation model of
diffusion
2. BT Success Implementation of a new Thesis
organisational capability (IT enabled
business process)
3. Conversion of Influenced by;
an innovation = Time to implement Thesis
into a new = Leadership Thesis
capability = Obsolescence rate (usage of the CAS Theory
new capability must be sustained
over time to be successful)
4.Communication | Overall effectiveness of Thesis
Structures
communication structures are
influenced by;
= APL
= Activity levels
= Group size
= Weak ties
This sector also models feedback
loops from;
= The new capability
= leadership
5. Adoption Rates
Models adoption rates based on the
interaction between potential, active
and former adopters. The sections also
models informal communication
(word of mouth) and weak ties.
= Bass diffusion
model
= Thesis
6. Communication
A base word of mouth level is
Bass model of
network modified by the interaction between diffusion
agents
7. Leadership Leadership in the communication = Thesis
network is influenced by the level of = CAS theory
management interest. Leadership has a
feedback effect on infectivity
(diffusion) rates and the
implementation of a new capability
8. Management
Interest
Management interest is influenced by
management conviction over time and
the feedback from the success of the
new capability
Bass diffusion
model
The system dynamics model developed from the above design parameters is set out in
graphical form , by sector in the following diagrams
w g Adoption of Innovation 8
New starts Combined adoption rate
Active time
Potential Adopters _/ Active Adopters
Former Adopters
adoption
lose interest quit
Quit % pa
Quit % pa
E=) 8
Base time to implement :
Saturation impact on implementation time
ime to implement
Leadership impact on implementation of New capability
New Capability
Noname 12
adopted %
Obsolescence rate % pa
Communication Structure 8
Impact of New Capability on Communication Activity
Average Path length
Effective Impact of Nq on Comm
Effective path length
Inpact of New Cap on path Length
base level contacts pm
New Capability
Contact Efficiency
Effect of Group size on Contact Efficiency
contacts pm
Effective Contact Efficiency
Group Size
base level of weak ties
impact of weak ties on formers participation
Connectedness
Saturation impact on implementation time
effective week ties
Impact of Leadership on wt
Volume of Communication Optimum no of contacts
Measures of Success 8
=
Staff Capability MOS
New Capability
Measure of Success
in MOS
Adoption Rate Calculations
effective directive adoption fract Probability Interact wih Active Total Adopted
directed adopt
Potential Adopters
tive Adopters FOxmer Adopters
q
Combined adoptign rate
opted %
Total Staff,
wom adopt
adoption from wom fract Probability interact with Former
New starts
Target Staff
Word of Mouth Effects 8
Volume of Communication
adoption from wom fract
Adoption contacts fract
worn axoption fract
Probability interact with Former
Former contacts
Leadership impact on infectivity & active time
base word of mouth infectivity
contacts pm
impact of weak ties on formers participation RraabIRy Interace we Active
Impacts of Leadership 8
Active time
Directive driven adopt fraction
effective directive adoption fract
Base Active time
Leadership impatt on infectivity & active time
Impact of Leadership on wt
Leadership impact on implementation of New capability
ow =) Feedback from Success of Implementation to Mgt Intetrest a
Instanteous Level of Mgt Interest
Base time to implement
apparent Level of Mgt Interest
Adj LoM! Mat Conviction
Adjustment Time
Delta Capability
Time for Mgt Interest to Decline
New Capability
Simulating BT Success using the Systems Dynamic Model
A series of sixteen simulations were run using ithink, in order to analyse the effect of
adjusting the key variables in the model;
Group size
Activity
Average path length
Overall connectedness
Leadership
Quit rate of adopters
The results are presented by the following graph that depicts five typical time series
representations of the model. The five simulations are:
1 C0'bS FS
Base case
Decrease in activity
Decrease activity and increased APL
Increased group size, decrease connectedness and decreased leadership
Increased activity, decreased A PL and increased leadership.
These simulations are graphed on the flowing table, with each time series being
numbered 1-5.
The results of these simulations are mapped against two success criteria, the
implementation of the new capability and the time taken to implement. The area under
the curve therefore represents the success of the BT project and can be quantified.
Variable Simulation | Simulation | Simulation | Simulation | Simulation
1 2 3 4 5
Group Size 50 50 50 100 50
Activity 200 100 100 200 300
APL" 5 0 9 5 1
Connectedness | 5 0 0 1 5
Leadership 5 0 0 1 10
Quit %age p.a. | 10 0 i) 1 10
Success 1049 640 281 218 898
Measure”
T Average Path Length
? Area under the curve
Other variables in the model held constant for all simulations;
= Base time to implement - 4 years
= Obsolescence rate per annum - 20%
= Base active time - 3
= Base word of mouth infectivity - 0.25
= Directive driven adoption fraction - 0.01
= Impact of communication on innovation - 0
Graphical Representation of BT Success Simulations
8 i:NewCapabilty 2: New Capability 3: New Capabilty 4: New Capability 5: New Capabilty
1: 100.004
bi |
bi
4 2 >:
_
| sacar | Sy
Za
1: 50.004 1
5
2
4 Se eee
aa —
$—L a
Ze 1
—_
1 0.00¢ =
0.00 12.00 24.00 36.00 48.00
8 BF ? Capability (Untitled) Months 4:21AM Wed, 21 Feb 2001
The model highlights the following key dynamics:
1. If one factor is removed or set at zero, the innovation takes a lot longer to
diffuse and successfully implement.
2; If two or more variables are removed or set at zero the innovation fails.
3. A saturation point for communication exists. Early take-up is experienced but
at a certain level of activity, the success falls off.
Conclusion
Results from communication network analysis can be incorporated into a diffusion of
innovation based system dynamics model for BT success. There is potential for the
model to assist the understanding of the varying communication structures in place to
support similar organisational change projects.
REFERENCES:
Bass, F. M. 1969, ‘A new product growth model for consumer durables’, Management
Science, 15 pp. 215-227.
Davenport, T.H. 1993, Process Innovation: Reengineering work through information
technology, Harvard Business Press, Boston, MA.
Hammer, M., & Champy, J. 1993, Reengineering the corporation: A manifesto for
business revolution, Harper Business Publications, New Y ork.
Holland, J. H. 1995, Hidden Order, Perseus Books, Reading, MA
Krebs, V. 1995, Inflow Reference Manual,
Krebs, V. 1996, ‘Visualizing Human Networks’ Release 1.0, Ester Dyson’s Monthly
Review, February 1996.
Krebs, V. 2000, ‘The Social Life of Routers Applying Knowledge of Human Networks to
the Design of Computer Networks’, The Internet Protocol Journal, (IPJ Issues), Dec.,
2000.
McElroy, M. 2000, ‘The New Knowledge Management’, Knowledge and Innovation:
Journal of the KMCI, 1(1) pp. 43-67
Nohria, N. & Eccles, R.G. 1992, Networks and organizations: Structure, form and
action, Harvard Business School Press, Boston.
O’Neill, P. & Sohal, A. S. 1999, ‘Business Process Reengineering: A review of recent
literature’, Technovation, 19 pp. 571-581
Rogers, E. M. 1995, Diffusion of Innovations 4" Ed., The Free Press, New York, NY.
Scott, J. 1996, ‘A Toolkit for Social Network Analysis’, ACTA Sociologica, 39 (2) pp. 211-
216.
Scott, J. 1991, Social network analysis: A handbook, Sage, Newbury Park, CA.
Stacey, R.D. 1996, Complexity and Creativity in Organizations, Barrett-Koehler, San
Francisco.
Sterman, J. D. 2000, Business Demands: Systems Thinking and Modeling for a Complex
World, Irwin McGraw-Hill, Boston, Massachusetts
Waldrop, M. W. 1993, Complexity: The Emerging Science at the Edge of Order and
Chaos, Penguin Group, London.
Wing, PJ. 2001, Business Transformation Success: Diffusion of Innovation and the
Pivotal role of Communication, PhD Thesis, Macquarie Graduate School of
Management, Sydney, Australia
Back to the Top
