long range planning

Long Range Planning 34 (2001) 287-308

www.lrpjournal.com

Hoshin Kanri: Implementing

the Catchball Process
Charles Tennant and Paul Roberts

Involving employees in strategy deployment is an active topic of strategic

management, but has been difficult to achieve in western companies. The
methodology known as Hoshin Kanri has proved an effective strategy deployment
process in Japan where it has been extensively applied for integrating strategy and
Total Quality Management (TQM), but its adoption in the West has been low, except
for a few innovatory companies. The application of Hoshin Kanri relies on a process
called catchball to gain consensus on the deployment of Hoshin targets and
measures in a team environment. This paper presents a process design based on an
adaptation of the Delphi technique for the effective implementation of catchball, to
reinforce the link between the corporate strategy and annual planning cycle. The
catchball process described was implemented at the Rover Group, a UK-based
automotive company, to develop the companys quality strategy based on Hoshin Kanri
principles.
c 2001 Elsevier Science Ltd. All rights reserved.

Introduction
One of the current active topics in strategic management is the
involvement of employees in the strategy deployment stages of
the process, particularly in western companies, which have
struggled with the concept of relating top management goals to
the daily work of employees. It is preferable if the top management goals are integrated through the formulation and development of policies and, by the involvement of a wider group of
managers through consensus. This should enable a higher level
of buy-in to the goals, improved understanding of the enablers
to achieving them, and the motivation to communicate them
into practical daily work processes. However, this has proved to
be a difficult process for western companies.
Many Japanese companies have already established an effective
strategy deployment process, known as Hoshin Kanri, which
attempts to integrate top management goals into daily oper0024-6301/01/$ - see front matter
c 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 2 4 - 6 3 0 1 ( 0 1 ) 0 0 0 3 9 - 5

Dr Charles Tennant is Principal

Fellow, Quality and Reliability,
in the Warwick Manufacturing
Group, School of Engineering,
University of Warwick. He
commenced his career as an
engineering apprentice at
Matrix Churchill Machine Tools,
and obtained a BSc (Hons)
engineering degree from
Coventry (Lanchester)
Polytechnic in 1985. He was
employed at the automotive
company Rover Group from
1985 to 1999, during which

time he worked in Powertrain

Engineering, Manufacturing,
Group Quality and latterly as
Chief Engineer of Land Rover
44 vehicles. He was awarded
an MSc degree in
Manufacturing Systems
Engineering in 1990 and an
Engineering Doctorate in 1998
from the University of Warwick.
His doctoral project was
Developing a Company Wide
Quality Strategy in the
Automotive Business. He is a
registered Chartered Engineer,
and Fellow of the Institute of
Mechanical Engineers. At
Warwick Manufacturing Group
he specialises in strategic
quality management and new
product introduction process
applications. Corresponding
address: Warwick
Manufacturing Group,
University of Warwick, Quality
and Reliability School of
Engineering, Coventry, CV4
7AL, UK. E-mail:
[email protected]
Paul A. B. Roberts is Principal
Fellow, Quality and Reliability,
in the Warwick Manufacturing
Group, School of Engineering,
University of Warwick. He was
employed in the Royal Air
Force from 1969 to 1990 in a
wide number of engineering
roles that included new
product introduction and
development of aerospace
equipment. He now heads a
team of specialists in quality
and reliability who lecture on
postgraduate programmes in
the UK and six countries
overseas. In addition to
lecturing and supervising
masters and doctoral students
he provides consultancy
support for partner companies
of Warwick Manufacturing

288

ations.1 This technique was originally developed in Japan from

the concept of Management by Objectives, and has been the subject of many English translations, which although similar, can
cause confusion in interpretation. The various translations
include policy deployment, policy control and management by policy. This paper does not intend to provide a full description of
Hoshin Kanri other than a brief overview. Instead, it explores
areas of difficulty for implementing the process in western companies. A particularly difficult phase of Hoshin Kanri is in the
implementation of a process known as catchball, which is used
to gain consensus on the deployment of Hoshin targets and measures, in a team environment known as cross-functional management. The difficulties are explored, and a process is presented
that was developed for and applied at the Rover Group, a UKbased automotive company, to develop the companys quality
strategy based on Hoshin Kanri principles.

Strategic planning issues

Corporate strategy is concerned with the direction a company
takes over time and with the use of its available resources; where
the best strategy for a company depends on current levels of
performance, particular strengths and weaknesses, and the
opportunities or threats in its particular environment. Some
ground rules for strategic planning are summarised as follows.2

Targets must be challenging but realistic so as to discourage

complacency;

Input from management must be encouraged;

Effective planning normally involves iteration;

Take a balanced and realistic view;

Commit sufficient time and thought to the process.

It is preferable that the line management who is to drive the

planning process also helps evolve it, with a stronger focus on
the most important issues, and a greater emphasis on implementation. Also, it is necessary to determine whether the strategic
planning process will trim ambitions to match the available
resources, or to leverage resources to reach seemingly unattainable goals, which has been previously termed Strategic Intent.3
The ascendance of Japanese companies has been seen as evidence
that western techniques for strategic planning have often failed,
where the former have tended to focus on quality, productivity
and teamwork rather than complex planning techniques.4 An
effective strategic planning process is an important enabler for
performance improvement in dynamic and complex industries,
but performance can be even higher when managers are authorised to make autonomous decisions and learn from their
actions.5

Hoshin Kanri

A particular approach for enabling the strategic planning process has been the adoption of Strategic Business Process Management, which requires the organisation to be conceptualised as a
series of business processes.6 Yet a fundamental question is what
should organisations do once they have developed a processbased architecture to ensure that the business process review does
not merely lead to a list of several dozen processes with little
further development? The crucial point is not the basic questions
relating to strategic planning, but rather the techniques and
organisational approaches which companies have traditionally
used to answer them. Therefore, to be effective, strategic planning must use a process, which is not separated from implementation, and involves a series of regular iterative reviews.

Hoshin Kanri
Overview
Hoshin Kanri is an organising framework for strategic management, which is concerned with the following four primary tasks.7

To provide a focus on corporate direction by setting, annually,

a few strategic priorities;

To align the strategic priorities with local plans and programmes;

To integrate the strategic priorities with daily management;

To provide a structured review of the progress of the strategic priorities.

Planning and deployment are critical elements of Hoshin Kanri,

which imply that the process of determining targets, the development of means to achieve the targets, and the deployment of
both are crucial for success. Hoshin planning principles are formulated around companies knowing what their customers will
want in five to ten years time, and understanding what needs
to be done to meet and exceed all expectations. The measurement system needs to be realistic, with a focus on process and
results and identification of what is important. Groups should
be aligned with decisions taken by people who have the necessary
information. Planning should be integrated with daily activity,
underpinned by good vertical and cross-functional communication. Finally, everyone in the organisation should be involved
with planning at local levels to ensure a significant buy-in to the
overall process. Figure 1 shows a model of the Hoshin planning
system, which can be compared with the Shewhart PlanDo
CheckAct (PDCA) continuous improvement cycle. A particularly crucial element within the Hoshin system is the Presidents annual diagnosis, which relies on a summary report of the
progress made, leading to either a reconfirmation or refinement
of the five-year vision and vital few hoshins or goals.

Long Range Planning, vol 34

2001

Group which are seeking

improvements to new product
introduction, manufacturing
and service processes and
research and development
strategies.

289

Figure 1. The Hoshin planning system

Adaptations
There have been a number of adaptations of Hoshin Kanri to
align the technique to western thinking and management
approaches. Examples include: Policy Deployment at AT&T, Managing for Results at Xerox, Goal Deployment at Exxon Chemical,
Policy Management at Florida Power and Light, Hoshin Planning
at Hewlett-Packard and Hoshin Kanri at Digital Equipment and
the FAIR model developed by Witcher et al.8 FAIR is an acronym, which relates to the PDCA cycle: Focus (act), Alignment
(plan), Integration (do) and Responsiveness (check).
One of the early innovators in policy deployment application
in the UK was Rover Group with the development of their fiveyear quality strategy through policy deployment.9 The Policy
Deployment project at Rover progressed through further iterations and refinements as an adaptation of Hoshin Kanri.10 This
paper will expand upon the catchball process developed for
implementation at Rover to ensure effective consensus with the
deployment of top management goals into working processes, as
operated daily by employees.

Research implications
Areas of difficulty
Although Hoshin Kanri has been extensively used in Japan for
more than 30 years as a methodology for integrating strategy and
Total Quality Management (TQM), its adoption in the West has
been low, except in a number of innovatory companies. This
may be due to confusion with the translation, or merely because
the approach incorporates a number of difficult phases which
require mastery. This is further compounded by the general lack
of reference to Hoshin Kanri in quality management textbooks,
other than superficial descriptions. Research in the UK has identified a particularly difficult nature of Hoshin Kanri implemen290

Hoshin Kanri

tation concerning the links between corporate strategy and annual

planning, where a process known as catchball, is recommended to provide an effective link.11 The research recommended developing methodologies to achieve a high degree
of co-ordination of Hoshin activities, with either normal daily
work processes, or specific projects. Therefore, the research question addressed within this paper is how can the consensus process
known as catchball, be effectively implemented in western companies to facilitate the Hoshin Kanri method, to translate top management goals into daily working?
Feedback into corporate strategy
Figure 2 shows the aforementioned FAIR model of Hoshin Kanri
aligned to the PDCA cycle. The Hoshin Kanri technique suggests
the application of catchball, to reinforce the link between corporate strategy and annual planning. In the FAIR model, this occurs
at the intersection between the focus (act) and alignment
(plan) phases. Previous research has identified that a weakness
in communication can often occur at this link, and that it is not
clear how feedback generated from the Hoshin cycle assists the
focus phase in up-dating corporate strategy.11 Furthermore, it
is suggested that in the UK, the absence of senior executive
involvement from a corporate centre actively participating in
annual audits, compounds the problems of aligning the corporate strategy with operational and quality issues.
Catchball process
The catchball process is necessary for successful implementation
of Hoshin Kanri in an environment of cross-functional management.12 Catchball is a term derived from a childrens ball game,
but instead of a ball, an idea is thrown around from person to
person. It is a critical element that requires continuous communication to ensure the development of appropriate targets and
means, and their deployment at all levels in the organisation.
Processes must be developed to ensure feedback in multi-directional horizons, which requires a company commitment to

Figure 2. The FAIR model of Hoshin Kanri

Long Range Planning, vol 34

2001

291

Having the right

balance of

employee involvement and continuous improvement. Crossfunctional management also relies on a refinement in the organisation of management relationships, to allow continual checking
of target and means throughout the Hoshin implementation
timescale. It is this iterative process of discussing and debating
plans and targets at each level until consensus is reached, along
with methods for meeting the goal, which ensures that the total
organisation is committed to the same goals.13 This paper will
address these fundamental research issues with the development
of a unique process, which has been designed to solve problems
associated with the implementation of catchball within the
Hoshin Kanri technique. In particular, it will present the concept
of involving managers in the formulation of policies by consensus, to effect the translation of top management goals into daily
working processes.

Catchball process design

management style
alone is not enough

Management style
When deciding on an appropriate process design for the consensus-building element, known as catchball within Hoshin Kanri,
it is necessary to consider the prevailing management style within
the organisation.14 It is suggested that the criteria by which
judgements are made, either by groups or individuals, are based
on the notion of shared mutual concepts and missions. This
implies a management style that is based on creative, flexible
and responsive principles. However, having the right balance of
management style alone is not enough. What is required is a
process for sharing ideas based on originality, coupled with the
benefits of group analysis and agreement whether that is in the
setting of Hoshin targets, or reviewing the achievement of targets
for incorporating within the next corporate strategy cycle.
Consensus building techniques
The research identified a number of consensus-building techniques for consideration within the implementation of the catchball process within Hoshin Kanri. These were reviewed in order
to optimise the development of the catchball process at Rover.
Ringi system
A particular consensus-building method known as the ringi system of decision-making is reputedly employed by more than 80
per cent of all Japanese industries.15 Rin means submitting a
proposal to ones superior and receiving their approval, and gi
means deliberations and decisions. The ringi system is characterised as a bottom-up method of decision-making, which has to
have unanimous approval of everyone in the system. Ringi allows
individuals to communicate a level of dissatisfaction without
actually halting the course of the decision. This approach is more
often applied in the decision-making processes within operations
rather than in the area of strategy. Therefore it could be adapted

292

Hoshin Kanri

in the West to provide a level of integration within Hoshin Kanri,

between the development of the strategic goals and the translation of these into real operational daily plans. The ringi
approach attempts to prevent the phenomenon known as
groupthink, which leads to a deterioration of mental efficiency,
reality testing, and moral judgement, resulting from in-group
pressures and an irrational tendency to follow instructions without question.16 The ringi system has many similarities to the process of catchball, and in fact reinforces the potential value of
Hoshin Kanri.
Social judgement
Social judgement theory assumes that an individual draws conclusions about unknown quantities or qualities on the basis of
available information, which can be applied as a concept for
achieving consensus in the development of strategic goals and
policies.17 Individuals may disagree in judgement, at least in part
because of the importance that each assigns to the available information. The earliest studies of social judgement compared the
accuracy of an individuals aggregated judgement, with a statistically aggregated judgement of several individuals. These were frequently called statistized groups because the individual members never met together. The most common aggregation rule was
a simple arithmetic mean, and the findings produced a consistent
pattern. Where the tasks had great intentional depth, the average
statistized group judgements were not shown to be significantly
better than the average of the individuals treated separately.
However, on tasks involving little intentional depth, the opposite
is true. The issue to be considered is what optimal sample size
from the population is required to achieve consensus? This was
carefully considered within the approach taken at Rover.
Delphi technique
The Delphi technique was originally developed by a US defence
research project in the 1950s, and has since been applied in the
Far East and Europe, primarily to develop long-range technological forecasts.18 The Delphi technique is defined as: A method for
structuring a group communication process so that the process is
effective in allowing a group of individuals, as a whole, to deal with
a complex problem. To accomplish this structured communication feedback must be provided of individual contributions
of information and knowledge, some assessment of the group
judgement or view, some opportunity for individuals to revise
views and some degree of anonymity for the individual
responses. The Delphi technique tends to be applied where there
is a recognised need to structure a group communication process
in order to obtain a useful result, as in the following examples:

The problem can benefit from subjective judgements on a collective basis;

The individuals needed to contribute may represent diverse backgrounds;

Long Range Planning, vol 34

2001

293

Catchball aims to
involve people as
team members rather
than individuals

More individuals are needed than can effectively interact in a

face-to-face exchange;

Time and cost make frequent group meetings infeasible;

The need to avoid dominance by quantity or strength of personality.

The Delphi technique requires a small monitor team who designs

a questionnaire that is sent to a larger respondent group. After
the questionnaire is returned the monitor team summarises the
results and using these develops a new questionnaire for the
respondent group. The respondent group is usually given at least
one opportunity to re-evaluate its original answers, based upon
an examination of the group response. An alternative approach
is often called a Delphi Conference which involves the use of a
computer to compile the group results in real-time. This
approach has the advantage of eliminating the delay caused at
the summary stage but does require that the characteristics of
the communication be well defined before the Delphi is undertaken. Some common reasons for the failure of a Delphi include:
poor selection of respondent group, too much use of jargon,
monitor teams imposing views and preconceptions, not allowing
for the contribution of other perspectives, poor techniques of
summarising, and not exploring disagreements so that discouraged dissenters drop out and artificial consensus is generated. It
is the task of the Delphi designer to minimise these problems
and balance the communication goals within the context of the
objective of the particular Delphi and the nature of the participants. The technique aims to develop consensus by avoiding
dominance by quantity or strength of personality, through questionnaire surveys analysed by a monitoring team. However,
employing statistised groups which do not meet during a Delphi
survey can lead to aggregated judgement, which does not
adequately take account of the best available judgement from the
most competent respondent, as demonstrated by the concept of
social judgement. Arriving at a balanced design for the communication structure is still very much an art.
Adaptation of Delphi
The key differences between social judgement and the Delphi
technique are shown in Table 1. The significant differences are
concerned with the absence of the individual respondents within
a Delphi survey actually meeting to exchange views and
develop consensus.
The catchball process aims to involve people as team members,
rather than individuals, based on the premise that groups perform at a level generally better than the competence of their
average members, but rarely as well as their most proficient
members. This team-based approach is usually achieved by application of the problem-solving tools, often associated with TQM.
Therefore, the consensus-building elements of the catchball process design need to be carefully considered.

294

Hoshin Kanri

Table 1. Delphi and social judgement techniques

Delphi technique

Social judgement

Absence of discussion among

group members

Presence of discussion among group

members

Focus on the evaluation and

feedback of outcomes

Focus on the information exchange

about the logic of judgement

Carefully restricted exchange of

information

Face-to-face encounters

Remote and private opinion

achieved through a series of
structured questionnaires

Involves use of discussion groups

The catchball process developed for Rover was based on an

adaptation of the Delphi technique to allow elements of group
interaction. Although it is clear that the effective implementation
of catchball within a Hoshin Kanri application is not absolutely
necessary, it was felt that it would improve the rigour of generating consensus at Rover. The application of the Delphi technique
within the strategy deployment process at Rover was innovative,
as research of UK companies has concluded that the Delphi technique only ranked 16th out of 23 tools and techniques often used
in strategic analysis.19

The catchball process at Rover group

Justification for single case study
The single case study example was selected because the principal
author was employed at Rover from 1985 to 1999, and was
actively involved in the development and implementation of the
catchball process. The work described in this paper represents
project work within the portfolio of the Engineering Doctorate
published by the principal author in 1998.20 The co-author was
initially employed at the Warwick Manufacturing Group in 1990,
to facilitate the introduction of quality and reliability tools and
methodologies at Rover and acted as academic supervisor to the
Engineering Doctorate. Although the narrow scope of single case
studies imposes limitations, in this case this is more than offset
by the depth of detail achieved, and the suitability of the research
context within UK manufacturing industry.
Rover Group
The UK-based automotive company Rover Group employing
around 40,000 on four manufacturing sites, one engineering
centre and a head office administration, had been through a
quality transformation by 1994 when it was acquired by BMW
of Germany. A TQM programme had been delivered in the late

Long Range Planning, vol 34

2001

295

A major process
review was conducted
to establish a
framework for the
quality strategy

1980s, but by the early 1990s was demonstrating signs of slow

down. The senior management team decided to develop a longrange quality strategy which could be used to direct the quality
efforts and measurement systems robustly. The process of creating the Rover Group Quality Strategy was initiated by the Quality Strategy Director (Dr Joe Cullen) in 1990, who appointed
the principal author as Quality Strategy Manager (Dr Charles
Tennant), the co-author (Paul Roberts) being actively involved
as a process facilitator. It was decided to develop the quality strategy based on the principles of Hoshin Kanri, in order to deploy
the companys vision and four strategic objectives, down to
working level processes. Rover had already been collaborating
with the Honda Motor Company for more than ten years and
was therefore familiar with Japanese management techniques.
Therefore, the case study will demonstrate a solution to the original research question: how can the consensus process known as
catchball, be effectively implemented in western companies to facilitate the Hoshin Kanri method, to translate top management goals
into daily working?
Initially a major company-wide process review was conducted
as a means to establish an agreed framework for the quality strategy. This was based on three broad steps and involved interviews
with more than 100 directors, managers, staff and front line customer service staff visits. The initial steps were determined as follows:

Identification of the main company macro-level processes;

Prioritise company processes against four key goals;

Agree company process for developing quality strategy milestones

based on catchball.

These steps were instrumental in developing the concept of the

quality strategy based on the principles of Hoshin Kanri.
Strategic policy deployment model
Before considering the design of a process to implement catchball, it is necessary to have an appropriate framework which
aligns the Hoshin targets with the business plan and organisation.
The design of the catchball process must enable this alignment
to be robustly deployed into business plans and management
action plans across the business. The review phase is a separate
stage to catchball, and must involve corporate staff who can use
the information in the annual corporate planning cycle. Therefore, following the creation of the quality strategy framework, a
model known as Strategic Policy Deployment, was designed to
relate Rovers vision and four strategic objectives with the companys key business processes, against a five-year time horizon,
as shown in Figure 3.20
It was decided to agree annual milestones for each of the nine
key business processes within the quality strategy. This would
296

Hoshin Kanri

Figure 3. Strategic policy deployment

relate to the concept of agreeing the means for achieving the four
strategic objectives, otherwise known as the vital few hoshins.
The milestone definition used was: an action or event whose
occurrence can be objectively verified and which will materially contribute to the improvement of the process. Therefore, a milestone
can be something which is carried out; such as re-engineer a subprocess, or achieve a result such as reaching a target level for
customer satisfaction. The main concept behind the milestone
plan was to distinguish between the hard and soft interdependencies often found in major projects. Hard dependencies are
the type typically encountered in sequential projects, such as
engineering programmes, whereas with soft dependencies it may
be possible to commence some tasks before the final training
programme is completed, as in the case of Statistical Process
Control (SPC). The extent to which any one activity must be
completed in a network of soft inter-dependent milestones for
the goal to be achieved depends on how far the other activities
have been completed. In this case, failure to achieve any one
milestone does not invalidate the plan: it simply imposes a higher
level of achievement as a minimum requirement on all other
milestones. This concept became known as geodesic planning,
which avoids defining specific inter-dependencies, instead relying
heavily on commitment to the vision.
Designing the catchball process
Agreeing the milestones for each year by consensus became a
major issue, for which a number of potential solutions were considered.

Long Range Planning, vol 34

2001

297

Quality strategy team research the major issues and formulate

the milestone plan based without achieving any real consensus,
or buy-in to the strategy;

Form small groups of senior executives and create the milestone

plan based on minor levels of consensus, and buy-in to the
strategy;

Involve a significant sample of the management population in a

company-wide consensus-building process, which would significantly aid buy-in to the strategy.

The first option was considered to be unsuitable, based on the

realisation that the quality strategy team would not be able to
identify all major issues intuitively, let alone accurately identify
effective remedial actions. The second option, although preferable to the first, would only be based on minor consensus, and
would not take sufficient account of the major company issues
and process dependencies. Therefore, the third option was selected based on the premise that as significant company resources
would be expended on the deployment of the quality strategy,
then it was justifiable to allocate commensurate levels of resource
and time to develop, clarify and communicate the strategy. Based
on the research into consensus-building techniques, an adaptation of the Delphi methodology was applied to design the
catchball process. The Rover Group Quality Strategy catchball
process is shown in Figure 4. A significant and novel feature of
this process, is the combination of individual one-to-one interviews, followed by the Delphi survey, review workshops and feedback to the key business process owners.
Implementation of the catchball process
The first step within the catchball process, was to identify and
interview a sample of up to 12 key stakeholders for each of the

Figure 4. The catchball process

298

Hoshin Kanri

nine key main business processes. These were essentially senior

executives who were process owners, internal customers or suppliers to the process as shown in Table 2. The interviews involved
a cross-section of company functions, thereby demonstrating
cross-functional management, as suggested within the Hoshin
Kanri approach.
The objective of the interviews was to seed the process with
initial ideas. They followed a consistent structure starting with
the key question: What is your vision of what the process should
look like in five years time? This vision was of the relevant key
business process being reviewed, rather than the overall company
vision. Once the process vision was established, the next subject
was the gap between the vision and the present situation. Understanding the gap then led to an identification of the milestones
necessary for achieving the vision. Only at the point during the
interview when a clear view of the vision, gap and milestones
was realised did the conversations move on to the issue of milestone timing.
The results of the interviews for each process were analysed
and converted into a milestone chart by using the affinity analysis
technique. The affinity analysis technique is one of the seven new
quality control tools which is designed to collect facts, opinions
and ideas about the unknown and unexplored areas that are in
a disorganised state.21 The technique is based on writing facts

Table 2. Business process owners, customers and suppliers

Process

Process owner

Customer

Supplier

New Product
Introduction

Project Directors

Sales and Marketing

Sales and Marketing

Manufacturing

Manufacturing Directors

Sales and Service

Sales Directors
Service Director

Sales and Marketing

Finance
Business Strategy Director
Logistics Directors
Product Directors
Sales Directors
Purchase Director
Manufacturing Directors
Logistics Directors
Product Directors

Business Planning

External Customers
Dealer Network
Product Directors
Manufacturing Directors
Personnel Director
All Company Managers and All Company Managers
Employees
Rover Learning Business
All Company Employees
All Company Managers and
Managing Director
Employees
Manufacturing Engineering Manufacturing Directors
Logistics Directors
Purchase Director
Business Strategy Director Executive Committee
All Company Managers

Logistics

Logistics Directors

Product Improvement

Product Directors

Management of People
Corporate Learning
Maintenance

Long Range Planning, vol 34

2001

Manufacturing Directors
Sales Directors
Manufacturing Directors

Sales Directors
Purchase Director
Sales Directors

299

(in this case proposed milestones) on to individual cards and

then sorting them into groups based on their affinity. It is a
particularly powerful team technique that depends on the right
half of the brain. That is the half that controls the ability to think
instantaneously, intuitively, emotionally and synthetically.
Initial milestones as agreed by the catchball process
Between nine and 12 milestones were identified for each process,
an example of which is shown for the New Product Introduction
process in Figure 5. It can be clearly seen that the milestones
represented a combination of both actions and events, where
milestone number one is an action, whereas milestone number
eight is an event.
The initial milestones were then issued in the form of a Delphi
survey to a much wider group of stakeholders of up to 150 on
some of the processes. The purpose of the survey was to gain
agreement on the process vision, and prioritise the milestones in
order of importance and achievement timing. Respondents to the
survey were subsequently invited to attend half-day workshops to
discuss the results of the analysis that had already been circulated.
This represented an adaptation of the Delphi technique, where
normally the respondents would never meet as a group. Circulating the results prior to the workshop made attendees aware of
the issues and allowed them to think about them. The output
of the workshops was again analysed using the affinity technique
to produce a proposed milestone plan for the process, which was
circulated to the original group of up to 12 senior stakeholders,
and finally presented to the Executive Committee for approval.
Once approved, the process milestones were established as company policy. In total, 89 milestones were agreed for the nine
macro-level business processes, by application of the catchball
process.

Figure 5. Initial milestones for new product introduction

300

Hoshin Kanri

Data collection
The case study was driven by primary data, as collected by the
principal author during the catchball process design and
implementation. The data collected during the initial interviews
was summarised into the Delphi questionnaire in the form of
initial milestones, which were summarised into date order simply
by mean and range from the survey respondents. This was carried out to provide points of discussion at the workshops, rather
than to determine analytically the planned date of achievement,
as it was the involvement at the workshops which provided the
main value in the consensus-building method within the catchball process.
Catchball process status and measures
Two main process measures were used within the catchball process. One was a holistic view of overall progress, the other an
individual process measure. The reason for instituting the measures was to ensure that the catchball process was implemented
with the rigour often associated with any normal company project. Both measures were presented in the form of simple charts,
which tracked the status for each process through the milestone
development stages, and were reported monthly to the Executive
Committee. Figure 6 shows an example of the catchball process
and measures of elapsed time, which identifies actual addedvalue process time and delay time, along with Delphi questionnaire response and group discussion attendance.
The elapsed time through the milestone development stages
was measured and recorded along with respondent rate to surveys and workshop attendees. These measures were critical for
two reasons. First, the initial application of the catchball process
for milestone development had taken 60 working days. This

Figure 6. Catchball Process and measures

Long Range Planning, vol 34

2001

301

The board members

themselves struggled
to reach consensus

would mean that it would take at least two years to get to the
first draft of the quality strategy. Second, it was essential to
ensure that the sample respondents achieved complete participation. By setting up the workshops up to three months in
advance and instituting rigorous follow-up procedures, it was
possible to reduce the total elapsed time through the milestone
development stage to 40 days, comprising 14 days process time
and 26 days delay time.
Figure 7 shows an example of the catchball process status
chart, which was used as a holistic measure to assess the
implementation of the catchball process against the nine key
business processes. In total approximately 700 managers were
involved in the development and agreement of the milestones
for each of the nine key business processes, and the total exercise
took 12 months and approximately 8,000 man-hours.
Communication process
The length of time required to operate the catchball process
across the nine key business processes, involving 700 managers,
and taking 12 months, required a high degree of commitment
and patience from Rover board of directors. Although having
secured the commitment early on in the process, the quality
strategy team encountered a major issue once the strategy had
been developed and concurred by the board. The main output
at this stage was the creation of a single A3 page format, which
encapsulated the complete strategy and 89 milestones, which represented an ideal communication media, as shown in Figure 8.
However, the board members themselves struggled to reach
consensus, with the novel idea to release the quality strategy to
the 3,500 managers within the company for the purpose of ongo-

Figure 7. Catchball process status

302

Hoshin Kanri

Figure 8. Rover Group quality strategy

ing deployment. The main concern surrounded confidentiality,

because while the strategy stated what needed to be achieved over
the next five years, it was also by nature an admission of what
had not been achieved to date. This was resolved by the then
Chief Executive (George Simpson), who in the true spirit of
Hoshin Kanri, declared the Rover Group Quality Strategy as the
companys business plan of significant importance to all company managers and authorised a communication cascade.
The quality strategy was eventually deployed into daily working processes across all company functions, became the subject
of regular review by self-assessment, and was further refined and
communicated annually over an eight-year cycle, from 1991 to
1999. It can be claimed therefore that the consensus process
known as catchball can be effectively implemented in western
companies to facilitate the Hoshin Kanri method, to translate
top management goals into daily working, as demonstrated by
the case study.

Discussion
This case study has addressed the fundamental issue of integrating top management goals to daily working, particularly in western organisations, where this has not been achieved with success
for a number of reasons. One of the main problems with western
organisations is that strategic planning is often considered, and
implemented, as a separate process from the operations side of
the business and the output subsequently imposed across the
organisation, as a series of projects and management action
plans. This tends to result in a lack of ownership across the company, as a major consequence of misunderstanding the strategic
objectives, and more importantly, of how to integrate them with
working level processes. Second, the strategic objectives often
become difficult to track in terms of progress results, due to the
aforementioned misunderstanding and lack of integration with
what actually happens in the day-to-day business of the company. A solution to this problem has been suggested as the adoption of the Japanese technique known as Hoshin Kanri which
attempts to integrate top management goals into daily oper-

Long Range Planning, vol 34

2001

303

ations. This is achieved by involving employees across the organisation in the strategic planning process through individual and
team contributions, in a non-threatening culture of consensus
building. Yet, although Hoshin Kanri has been successfully
applied in many Japanese companies, its adoption in the West
is considered as low.
A review of the main issues surrounding the difficulties in the
application of Hoshin Kanri within western organisations concluded that one particular area of difficulty concerned the links
between corporate strategy and annual planning. In order to achieve a high degree of co-ordination of Hoshin activities, with
either normal daily work processes or specific projects, a consensus-building process known as catchball is needed. Therefore, the
research question addressed within this paper was how can the
consensus process known as catchball, be effectively implemented in
western companies to facilitate the Hoshin Kanri method, to translate top management goals into daily working? This was
developed following a conclusion drawn at Rover that Hoshin
Kanri had already been demonstrated by Japanese companies to
be an effective strategy deployment technique. In particular, it
was felt that Hoshin Kanri would achieve effective involvement
of managers, generally in strategy development and deployment
throughout the organisation. However, it was concluded that to
apply catchball effectively within the cultural constraints of a
major western organisation a unique process would need to be
designed, rather than be expected to occur naturally.
A catchball process was designed for application at Rover as
a means for agreeing improvement milestones for the companys
quality strategy. The process was developed using a novel combination of the Delphi technique to achieve individual contributions, and consensus-building workshops based on group
judgement approaches, by considering the outputs of the Delphi
surveys. The catchball process involved approximately 700 managers totalling 8,000 man-hours contribution over a 12-month
period, to agree the initial 89 milestones contained within the
Rover Group Quality Strategy. The milestones were agreed over
a five-year horizon covering the period 1991 to 1995, and were
aligned to the nine key business process visions, four strategic
objectives, and the overall corporate vision. This delivered a strategic plan, which conformed to the principles of Hoshin Kanri,
and was communicated across the company in a one-page A3
sized format.
An early learning point was that managers initially struggled
with the concept of thinking more than one year out, and
required facilitation throughout the interviews to develop their
strategic thinking. However on most of the key processes, the
basic issues started to crystallise after about four to five interviews, which could then be used as input on subsequent interviews. However care was taken to ensure that the interviewees
were not stifled by previous comments, and therefore were
allowed free thought during the interview. Another area of difficulty was to prevent what became known as high guts to brains
304

Hoshin Kanri

ratio, where managers would suggest unachievable milestone

dates within the five-year time horizon in their responses to the
Delphi survey. This was based on misguided commitment to the
process, and was resolved by running group workshops to flush
out the main issues. It was this modification to the Delphi technique that created rigorous levels of consensus within the management sample. A second learning point was the requirement
for the development of an effective cascade process for the strategy, incorporating a results measurement system. This was achieved with the creation of a model known as Strategic Policy
Deployment, which explained how company functions should
communicate the strategy, by developing local management
action plans integrated with daily working processes and use selfassessment measures.
The solution adopted for implementing the catchball process
at Rover was based on two main areas of existing knowledge:
the Delphi technique, and group judgement methodologies.
These were combined in a novel process, which allowed the company to overcome cultural resistances to the notion of catchball
for achieving consensus by involving employees in the strategic
planning process. In particular, the outputs from the Delphi surveys were used to create specific data for discussion at the group
workshops. It was found that once this data was created, it was
not necessary to use the Delphi technique on subsequent annual
iterations of the quality strategy, as a critical mass of understanding had already been initially achieved. Therefore, only the group
judgement elements were retained, based on process workshops
to review progress against the milestones, and refine or agree
new milestones. In this way, the Rover Group Quality Strategy
became self-sustaining over a period of nine years, up to 1999.
By using the results of the self-assessment reviews, the process
workshops were able to build upon past achievements.
However, inconsistencies were found in the effectiveness of
the strategy deployment between the operations side of the company and the overall business planning process. The operations
side used a fractal approach, which encouraged consistent application across the areas of engineering, manufacturing and sales
etc. The business planning did not, which caused deficiencies in
the creation and commitment to the product strategy, inadequate
implementation, and eventual collapse of the company through
brand confusion and inappropriate marketing strategies. This has
led to the decision by BMW to break up the company by selling
the Land Rover side to Ford Motor and the Rover Cars to the
MG Rover Company (formerly known as the Phoenix
consortium).
This case study highlights implications for both current work,
and future research. The results have demonstrated that the
catchball process designed and implemented at Rover Group,
achieves three main outcomes in the form of retention of: individual thinking, reinforcement of individual thinking by wider
contribution, and a greater acceptance of the planning outcomes.
It can be concluded therefore, that the initial commitment and

Long Range Planning, vol 34

2001

A novel process
allowed the company
to overcome
resistances to the
notion of catchball

305

resources expended were beneficial, in that the Rover Group

Quality Strategy was accepted across the company at all levels,
and was in place for nearly ten years. Areas for future research
opportunities involve the development of intranet-based data
collection and analysis, within the Delphi and group judgement
elements of the catchball process. This should be carried out with
the specific aim of shortening the cycle time, reducing the number of man-hours involved, and further increasing levels of
employee involvement in the strategic planning process.

Conclusions
1 Research has demonstrated that effective application of
Hoshin Kanri requires mastery of the concept of a process
known as catchball for building consensus, to translate top
management goals into effective policies and working level
processes.
2 A catchball process has been designed based on an adaptation
of the Delphi technique by incorporating group judgement
methods, which was aimed at achieving consensus from a
large sample size. This was implemented at Rover by involving
approximately 700 managers over a period of 12 months
(8,000 man hours), in the development and agreement of
milestones to achieve policy deployment of Hoshin Kanri targets. The application of the Delphi technique within a strategic
planning and deployment phase represents innovation in
western organisations.
3 The solution adopted for implementing the catchball process
at Rover allowed the company to overcome cultural resistances to the notion of catchball for achieving consensus. This
was achieved by involving employees in the strategic planning
process, applying the outputs from the Delphi surveys to create specific data for discussion at the group workshops.
4 The one-to-one interview stage within the catchball process
assisted in getting managers to think more than one year out,
by iterating outputs from previous interviews as inputs to subsequent interviews. Measures were applied to manage the
implementation of the catchball process as a project and
improve efficiency. Also, the workshop stage within the catchball process, which represented an adaptation of the Delphi
technique, was facilitated to ensure that achievable dates for
each individual milestone within the five-year time horizon
were agreed by consensus.
5 It was not necessary to use the Delphi technique on subsequent annual iterations of the quality strategy, as a critical
mass of understanding had already been achieved. Therefore,
the group judgement elements were retained, based on process
workshops to review progress against the milestones from the
306

Hoshin Kanri

self-assessment reviews, and refine or agree new milestones,

thus building upon past achievements.
6 Inconsistencies in the effectiveness of the strategy deployment
between the operations side of the company, and the overall
business planning process were identified, which led to a
poorly developed product strategy, inadequate implementation, and eventual collapse of the company.
7 The results have demonstrated that the catchball process
designed justified the initial commitment and resources
expended for the development of the Rover Group Quality
Strategy, which was accepted across the company at all levels,
and was in place for nearly ten years.

This work was undertaken as

part of the UK Government
EPSRC sponsored Engineering
Doctorate Programme at the
University of Warwick and Rover
Group between 1993 and 1998.
The authors appreciate the
constructive comments received
from the reviewers during the
draft stages of this paper.

8 The catchball process presented as applied at Rover represents

a leading edge approach to involving managers in strategy
development and deployment throughout the organisation.
The case study also demonstrates that Hoshin Kanri can be
applied in western companies as a means for translating top
management goals into daily working.
9 Future research opportunities involve the development of
intranet-based data collection and analysis, within the Delphi
and group judgement elements of the catchball process. The
specific aims should be: to shorten the cycle time, reduce the
number of man-hours involved, and increase levels of
employee involvement in the strategic planning process.

References
1. Y. Akao, Hoshin KanriPolicy Deployment For Successful
TQM, Productivity Press, Cambridge, MA (1991).
2. B. Houlden, Understanding Company Strategy, Blackwell,
Oxford (1990).
3. G. Hamel and C. K. Prahalad, Strategic intent, Harvard Business Review, June (1989).
4. M. E. Porter, Corporate strategy: the state of strategic thinking, The Economist, 23rd May (1989).
5. T. J. Anderson, Strategic planning, autonomous actions and
corporate performance, Long Range Planning 33, 184200
(2000).
6. C. Armistead, J. P. Pritchard and S. Machin, Strategic business process management for organisational effectiveness,
Long Range Planning 32(1), 96106 (1999).
7. B. Witcher and R. Butterworth, Hoshin Kanri: how xerox
manages, Long Range Planning 32(3), 323332 (1999).
8. B. Witcher and R. Butterworth, What is Hoshin Kanri? A
Review, ESRC, Economic and Social Research Council, UK
(1999).
9. J. M. Cullen, Visualising improvement, The TQM Magazine,

Long Range Planning, vol 34

2001

307

10.

11.

12.
13.

14.

15.
16.
17.

18.
19.
20.

21.

308

The International Bi-monthly for Total Quality Management,

August (1993).
C. Tennant and P.A.B. Roberts, Using Hoshin Kanri for
strategy deployment, International Journal of Manufacturing
Technology and Management 2(17), 517531 (2000).
B. Witcher and R. Butterworth, The Hoshin Kanri method,
the quality journey, The Second World Congress for Total
Quality Management, Sheffield Hallam University, UK
(1997).
G. H. Watson, Strategic Benchmarking, Wiley, New York
(1993).
R. G. Lee and B. G. Dale, Policy deployment: and examination of the theory, International Journal of Quality and
Reliability Management 15(5), 520540 (1998).
S. N. Daigaku, Vision Management: Translating Strategy into
Action (Fred Czupryna, Trans.), Productivity Press, Portland,
USA (1992).
L. J. Irving, Crucial Decisions: Leadership in Policy Making
and Crisis Management, Freepress, New York (1989).
M. Y. Yoshino, Japans Managerial System, Tradition and
Innovation, MIT Press, Cambridge, MA (1976).
J. Rohrbaugh, Improving the quality of group judgement:
social judgement analysis and the Delphi technique, Organisational Behaviour and Human Performance (24), 7392
(1979).
M. Linstone and M. Turoff, The Delphi MethodTechniques
and Applications, Addison-Wesley, London (1975).
K. W. Glaister and J. R. Falshaw, Strategic planning: still
going strong?, Long Range Planning 32(1), 107116 (1999).
C. Tennant, Deployment of a company wide quality strategy
in the automotive business, Executive summary, Engineering
Doctorate (Eng.D), University of Warwick, UK (1998).
H. S. Gitlow, Planning for Quality Productivity and Competitive Position, Dow Jones-Irwin (1990).

Hoshin Kanri