IMPLEMENTING QUALITY INITIATIVES IN INFORMATION SYSTEMS
James A. Ward, PMP
Information Systems organizations face
unique problems in implementing quality improvement initiatives. Even in corporations that have made great
strides in implementing quality improvement, Information Systems has not always
kept pace. We will discuss how the
principles of quality improvement can be modified and applied to Information
Systems.
MAJOR PROBLEMS FACED BY INFORMATION SYSTEMS ORGANIZATIONS
Meeting the customer's requirements is
at the core of all quality improvement efforts. In information systems, discovering the true customer
requirements is often the most difficult part of the job. We will see how quality improvement can
offer specific techniques to make this job easier.
Continuous improvement of processes is
essential to achieving quality. Most
Information Systems organizations do not even have defined processes. A minimum set of stable and repeatable
processes must be defined, implemented, standardized, measured and controlled
for quality improvement to be sustainable.
THE CORE MESSAGES OF QUALITY IMPROVEMENT
First, let's summarize what the quality
movement means. We can define the four
basic concepts and then illustrate how they can be applied to Information
Systems to solve the problems we have cited.
1. Intense
Focus on Customers and Their Satisfaction
You must identify, measure and
"design in" the product/service attributes that the customer cares
about. Customer satisfaction must be
continuously monitored and measured.
The organization must make sure that everybody knows their
customers, both external and internal.
Make sure everybody can "see" the ultimate customer
using your products and services.
2. Use
Problem Solving Methods and Measures to Achieve Continuous Improvement
Define your processes. Systematically apply analysis and problem
solving to improve processes. Never be
satisfied with the process. "If it
ain't broke, FIX IT." Continuously
apply and repeat these steps.
3. Use
Teams and Teamwork, Encourage Employee Involvement
Use teams to maximize problem solving
power, support and commitment/buy in.
Form teams around your work processes:
Functional teams - to address everything the department does. Cross-functional teams - to break down
barriers between department. System
teams - spanning vendors, company personnel and customers.
4. Manage
a Process That Encourages the Core Messages
Management must lead the quality
process, creating a climate that allows ideas and information to flow freely
and without fear. Manage using
participative methods to maximize input, ideas and commitment/buy in. Management should expect that everyone
understands quality and has a personal responsibility for quality. Management must recognize and reward quality
work.
USING QUALITY IMPROVEMENT TO ADDRESS INFORMATION SYSTEMS
PROBLEMS
The key concepts of quality improvement
can provide significant assistance in addressing and solving the problems we
all face in Information Systems.
Meeting Customer Requirements
It is essential that we address those
things which the customer cares about.
We must define and measure the real products and services that we
provide. Our products are essentially
"information." They are not
technology, lines of code, function points, software or even systems. Measure what the customer cares about.
Make sure we know who the true customer
is. Don't confuse "customer"
or “client” with "user."
Often both we and the user are working for the customer or client. We satisfy our customer by maximizing the
benefits of information systems to the whole organization.
I recently read an article by officers
of a "Big 5" consulting firm discussing how to "manage"
customer expectations. In
"manage" you can read "limit" or "control." This to me is exactly the wrong
approach. To be ultimately successful,
we must continuously exceed customer expectations while at the same time
continuously raising those expectations.
Why do we have such problems in meeting
customer requirements and how can quality improvement help us? We have problems because of the nature of
the working relationships with our customers.
The customer (or user) does not always have the ability to envision a
new information system until it can be seen, felt, touched and used.
On the other hand, system developers often lack sufficient knowledge of
the business to understand the context in which the system must be used and how
it will work in the business environment.
Consequently, there is a real communications problem in requirements
definition.
I once had a programmer tell me that
you never learn the requirements until you're into testing, so the thing to do
is code something and see what happens.
Most of us have heard the admonition that we can never give the users
what they need until after we have given them what they want.
In actual practice, what happens in
organizations with advanced quality improvement programs, is that they move
beyond reliance on the customer's perceived requirements to identifying
products and services which the customer truly wants and needs. They are not dependent on asking the
customer.
Continuous Improvement of Processes
Before we can implement continuous
process improvement, we must establish some basic processes for doing
information systems work. Quality guru
Philip Crosby said that quality improvement is a system of prevention. We must do the job right the first time and
every time. The way we accomplish this
is to focus on continuous improvement of processes through simplification and
reduction of variability. Processes
must be stable, repeatable, visible and measurable. We cannot control what we can't measure. We must set a standard of zero defects.
However, most Information Systems
organizations lack even a minimum set of processes. Motorola, one of the first organizations to embark on the quality
journey, found that the system development process must be nudged from an individual
art form to a measurable process before it can be rigorously controlled. While it is true, unlike most manufacturing
operations, that we never develop exactly the same system twice, we can and
should utilize the same set of processes to develop all our systems. The
absolute minimum set of processes that must be defined and implemented in an
information systems organization are:
A System
Development Life Cycle Methodology
A Project Management Methodology
A Quality
Assurance and Testing Methodology to include a system of Walkthroughs,
Inspections and Technical Reviews
Without these
processes there is no basis for continuous improvement.
IMPLEMENTING QUALITY IMPROVEMENT IN INFORMATION SYSTEMS
The concepts and methods of quality improvement
have been part of the conventional business wisdom in America since at least
the mid 1980s. Until recently, there
has been an insufficient body of knowledge and experience to understand the
unique application of these disciplines to Information Systems.
The Five Stages of Quality
In his book, Quality Is Free,
Philip Crosby defined something he called the "Quality Maturity
Grid." He classified management
attitudes about quality into five categories and suggested that improvement
happens as management "matures" from one category to another. These categories are Uncertainty, Awakening,
Enlightenment, Wisdom and Certainty.
Watts Humphrey of the Software
Engineering Institute (SEI) applied this same thinking to the types of
processes one finds in an Information Systems organization. He defined five categories of processes
which conform to Crosby's and codified these in the SEI’s Capability Maturity
Model (CMM). These are Initial,
Repeatable, Defined, Managed and Optimized.
Subsequently, Gerald M. Weinberg has
taken this work and related it to the cultural patterns that one finds in these
organizations. He called the five
stages Variable, Routine, Steering, Anticipating and Congruent.
This work is important as it gives us a
basis for identifying organizational processes and formulating plans to
increase quality. Our goal is to
identify the level of management attitudes, types of processes and cultural
patterns within our organization and formulate a plan for moving to the next
higher level.
Level 1 - Variable, Uncertainty, Initial
Most information systems organizations,
Humphrey and Weinberg estimate about 80 per cent, operate at this level. Quality depends solely on the
individual. There is no dependability
or predictability of schedules across projects because individual differences
are so great. Management has no
knowledge of quality as a management tool.
Further, there is no knowledge of management as a systems development
tool. The environment is ad hoc and
often chaotic.
A typical attitude about quality is
that there is no idea as to why there are quality problems. The only measurements of quality are whether
or not the system works and the personal working relationships between
developers and users. Some work may be
excellent, some may be bizarre. It is
totally dependent on the individual.
What we find in organizations like this
is the impossibility of attempting projects of any appreciable size. A one to two man project of duration of less
than one year is about the largest that can be attempted in this environment
with any reasonable hope of success.
The organization lacks the ability to coordinate the efforts that cannot
be easily handled by one or two individuals.
Level 2 - Awakening, Repeatable, Routine
At this level it becomes understood by
management that projects require coordination.
Some procedures are instituted, often informal. One of the first things that is instituted
is time reporting. An attempt is made
to provide schedules. These things are
done by management fiat. However, there
is little understanding of the dynamics of the process and no visibility or
feedback mechanisms for measurement and control. The organization often loses control when projects don't follow
the proscribed routine. The process is
still dependent on the individuals actually performing or managing the
work. It is estimated that about 15 per
cent of Information Systems organizations operate at this level.
A typical attitude about quality in
these organizations is to ignore problems and hope they will go away. If management doesn't recognize quality
problems they won't have to deal with them.
Personnel only go through the motions
of following procedures, with no real understanding of them. Procedures are quickly abandoned when things
go wrong and the organization reverts to Level 1 performance. Schedules are regarded as mere "Wish
Lists" with no bearing on reality.
Level 3 - Enlightenment, Defined, Steering
This is the minimum level required for
sustained quality improvement. It is
also the minimum level at which productivity improvement tools and
methodologies can be effectively used.
Procedures are defined and followed, even in a crisis. Management understands the process. There is feedback, early and often. The process is stable and controlled. Larger projects, requiring coordination of
several individuals, can now be undertaken with some assurance of success. Maybe five per cent of all information
systems organizations consistently operate at this level or higher.
The organization believes that through
commitment to quality improvement, problems are being identified and
solved. Initial attempts at measurement
are made by querying customer response, but this is not generally done
systematically. The organization enjoys
consistent success in meeting commitments.
The focus is on controlling the product, at key steps within the
process.
Level 4 - Wisdom, Managed, Anticipating
There is an increased understanding of
procedures, which are followed uniformly.
Comprehensive process measurements and analysis of processes has been
implemented. This level is where
significant quality improvement really begins.
Few organizations have achieved consistent operation at this level.
Defect prevention - doing it right the
first time - is now a routine part of the operation. The organization achieves consistent success, even on the most
ambitious projects. They are constantly
improving the quality of all products by improving the quality of the processes
used in creating those products. There
is a high visibility of the processes.
This is the major difference between levels 3 and 4. Level 3 attempts to control the quality of
the product. Level 4 attempts to
control the quality of all products by controlling the quality of the
processes.
Level 5 - Certainty, Optimized, Congruent
All personnel understand and follow
procedures, which everyone is involved in improving at all times. Sophisticated and automated measurement of
quality is done on a continuous basis.
At this time, this level is basically theoretical for information
systems organizations, although a few organizations have achieved CMM level 5
process maturity.
These organizations do not have quality
problems and they know why they don't.
Quality is measured by mean time to customer failure, usually many
years. Customers can bet their life on
quality. The environment is totally
responsive to the customer.
HOW TO GO ABOUT IT
In implementing quality improvement in
Information Systems, regardless of the current quality maturity level of the
organization, there are certain specific steps which must be taken.
Define Processes
The first thing that must be done is to
establish a minimum set of defined, repeatable, measurable and controllable
processes. These processes must include
a formal System Development Life Cycle Methodology and a formal Project
Management Methodology.
Establish Standards
The organization must establish
standards of performance for every task contained in the System Development
Life Cycle and the Project Management Methodology.
Implement Quality Assurance
The organization must implement a
formal System Quality Assurance and Testing Life Cycle and integrate it with
the System Development Life Cycle. A
system of Walkthroughs, Inspections and Technical Reviews must be implemented
based on established standards.
Implement Techniques and Methodologies
It makes little difference what
specific tools and techniques are used, quality improvement benefits can be
realized with any of these. The
organization may wish to use such things as Structured Techniques, Rapid
Analysis, Joint Application Development, Prototyping, Object Oriented
Techniques, etc. Some of these
techniques may be implemented at the time standards are established if the organization
chooses to standardize around a particular approach to system development.
Automate the Processes
At some point in the process,
automation will be required to achieve truly significant improvements. Automation will further standardize the process,
reduce variability, enable sophisticated measurement and control.
CASE tools, Object Oriented Development
and automated project planning and accounting tools can enable tremendous
quality and productivity improvements, but only after the groundwork has been
laid. Automation will not achieve
positive results until stable processes are defined.
Measure Processes
Sophisticated and automated measurement
of product and process quality is necessary to achieve the highest levels of
quality improvement. You can't control
what you can't measure, and the better you are able to measure what you do the
better you are able to control it.
Again, it must be remembered that you should measure the things that the
customer cares about. A sophisticated
measurement system that measures the wrong things will have a negative effect
on quality. what you measure is what
you get.
PROBLEMS THAT WILL BE ENCOUNTERED
How can we anticipate and overcome the
problems of implementing quality improvement in Information Systems.
Dealing With the Impetus For Change
All organizations are resistant to
change. An organization will only
undertake change when it can no longer meet the demands placed on it in the
current mode of operation. Demands will come from customers or potential
customers, the problems that the organization is attempting to solve and by the
threat of competition. In the short run
it is possible to resist change by trading one set of demands for another, but
only in the short run.
Difficulties Encountered
Management may be satisfied with the
current mode of operation, including quality and productivity.
Management may fear losing the current
level of quality and productivity in an attempt to do better.
Management lacks understanding of other
ways of managing the information systems function. Nobody will adopt a technique they don't know about.
Management must always deal with the
invisibility of their own culture.
Errors
There are only two reasons for errors
that result in poor quality: lack of
knowledge, and lack of attention. Lack
of proper tools is a lack of knowledge or attention on someone's part. Failure to provide adequate training and
direction is one the most common signs of lack of management commitment to
"Total Quality."
The Major Impediment
The single greatest impediment to
quality is the top Information Systems executive. There is simply a lack of understanding and commitment at this
level. Information Systems executives
must change their attitudes about standards.
However, when the Information Systems executive converts, they become
the greatest force behind quality.
Quality is, after all, a management problem.
Note that fully two thirds of all
quality improvement efforts fail to realize significant results. The reason is always the same - lack of
management understanding, commitment and involvement. The process is never self-sustaining.
ABOUT THE AUTHOR
James A. Ward is an
independent management consultant specializing in System Development Project
Management and implementation of Quality and Process Improvement Initiatives in
Information Systems. Mr. Ward has over
thirty years of experience in information systems management and consulting. He holds an MBA in Finance and Business
Policy from the University of Chicago and a Bachelor's Degree in Economics and
Mathematics from the University of Minnesota. He holds a PMP certification from
PMI.
Mr. Ward resides in
Jacksonville, Florida. He can be
contacted at (904) 273-8777 or via e-mail at soozward@earthlink.net. Visit his web
site at http://www.JamesAWard.com.