ACURIL: Association of Caribbean University, Research and
Institutional Libraries. XXV Conference, 1995, San Juan, Puerto Rico
Designing Information Systems
meet our Clients' Needs.
Professor Tom Wilson
Department of Information Studies
That we should understand the needs of our clients in order to
deliver effective information services is such an obvious statement
that it is almost embarrassing to make it. However, general
observation, the reports of students who have been subject to the
vagaries of university library systems, and consultancy work in
organizations have convinced me that it is a statement well worth
repeating. Merely repeating the statement, however, is not of very
much assistance to librarians and information workers who are subject,
increasingly, to demands for more cost-effective (i.e., cheaper)
services, and to demands for greater relevance of their work to the
needs of the organization. In addition to rhetoric, they need help -
help to understand the nature of information needs, how those needs are
generated by the work that people do in organizations, the behaviour
involved in gaining access to the needed information, and the role of
formal library and information services in delivering the information.
This paper is an attempt to provide some guidance - in so far as that is
possible in the course of a conference paper.
The term 'information need' has been a troublesome one: writers
have debated the difference between 'needs' and 'wants', between
'perceived needs' and 'expressed needs', and so on; and, of course,
librarians and information workers have always been willing to decide
that they know what the user needs, better than the user. However, very
little emerges from this debate to guide our actions - in the end, we
have to try to anticipate demands upon our services and, therefore, we
have to act as if we understand the needs of the information user.
Some guidance through the fog of information needs is possible if
we understand the nature of the organizations we work in, if we
acknowledge that the organization itself (in terms of its goals and
objectives) has needs for information; and if we understand that
personal needs arise out of the roles people undertake in organizations.
The nature of organizations
The first point to be made is that all organizations are information
and communication systems: some more so than others, of course. For
example, a media company, such as a newspaper, deals almost entirely in
terms of information flows (from the wire services and from its own
journalists and other writers) and the subsequent organization and
re-presentation of that information. A bureaucracy, such as a
government department, is also almost entirely an information
processing and communication organization: everything is document,
records and files abound, filing systems and records management are
central to the organization. Even a manufacturing company is mainly
concerned, at head office level, with the information flows about
production and market performance.
The question for the information manager or librarian, therefore, is
not whether information is important in the organization, but how
information intensive are its operations
Organizational information needs
If we accept the information-processing model of organizations it
follows that organizations will have information needs, if they are to
satisfy their goals and objectives. Recently, I have been exploring
the use of Porter's model of the 'value chain' in identifying the
information-critical areas on an organization's activity. This model
suggests that different areas of the value chain will be of greater or
lesser importance to a company and that, consequently, information will
have different levels of importance.
One example is the pharmaceutical industry: clearly, except for
companies that manufacture only generic drugs (which are those for
which the patents have run out), the research and development function
is critical - conse quently, all such companies have major information
services departments, as well as well-defined procedures for drug
testing and registration. In another case, the publishing industry,
one of my Ph.D. students has determined that the marketing function is
the most critical area of the value chain, but that the information
consequences of this have not been understood by the organiz ation
So - value chain analysis of your organization is one way to
identify where the information needs are most likely to lie and, most
importantly, where your efforts can be of maximum benefit to the
Work in the organization
We also need to understand the work that people do in the
organizations we serve. Many years ago, when I worked in an
industrial research organization as an information manager, the
Director of the organization told me that he expected me to spend
25-30% of my time walking around the laboratories, talking to people,
discovering what they were doing and understanding the nature of their
"Walking the organization" is a well-practised management strategy
for finding out what is going on, but I know few librarians who
practice it and fewer still who require their staffs to practice it.
Most stay metaphorically chained to their desks and are reluctant get
out and about in this way, in case they are viewed as wasting their
time. Instead, they ask about ways of finding out about users needs
through questionnaire design and surveys, or through analysing
departmental records of information use.
Such tools are useful, but, in my opinion, they are no real
substitute for getting out and talking to people.
Once you do get out, what are you likely to find? Many things:
you should already know how the organization is structured and what it
does, in general terms. You might also know of its overall strategy,
whether it is a public sector strategy, or a private sector strategy -
you should know what it is trying to do. And, if you don't, finding
someone to ask is a good place to start.
Observing what is going on, however, and talking to people about
their work, will reveal a great deal more: how much work is done at
the desk, how much at meetings, how much outside the organization.
How information is brought into the workplace (other than through your
information service), what external links to information sources exist
and how they are used; what current problems exist in the department
or section, or in the individual's work and those problems are being
tackled. Perhaps, in these conversations, you too will be able to
made a contribution - drawing attention to untapped information
resources, agreeing to explore the information aspects of a current
problem and report back, or simply by knowing, by chance, the answer to
a current problem.
Through interaction of this kind you become a "well-informed
citizen" of the organization, one who can act as an effective
information intermediary - not only between people and information
sources, but between one person and another. I once saved the
organization I worked for a considerable amount of money by discovering
that two people in two different departments were both working on the
same problem in complete ignorance of one another's existence - until I
provided the link.
Personal information needs
We generally assume that the work people do gives rise to cognitive
inform ation needs - that is, a need to fill gaps in knowledge. The
knowledge may be trivial - the address of a supplier, for example; or
it may be fundamental to a policy issue or a scientific problem - the
current number of homeless persons in a local authority area, or a
method for analysing a hazardous substance. And this is the kind of
information need for which information services are generally set up.
Information needs of this kind can be 'unearthed' during our walks
around the organization, or we can discover them in general terms
through surveys using standardized lists of topic areas. However, we
can also design our systems so that they deliver information about the
users and their information choices: for example, automated loans
systems can provide a fund of information of this kind; monitoring
inter-library loan requests, analysing photocopying forms, in fact
regular monitoring of all information system/service activity can
provide us with information that raises questions that we can ultimately
ask of the user and from which we can derive indicators for future
pro-active actions on the part of the information service.
However, information may also be used to satisfy affective or
emotional needs. I discovered this some years ago when carrying out
research on communication in social welfare departments in the U.K.
People have different kinds of affective needs in their jobs: they
need to feel competent and assured in their roles; they may feel a need
for achievement or ambition; at the more pathological end of the
spectrum they may feel a need to dominate and control through the
possession of greater knowledge.
These affective needs are more difficult to determine, but they can
come to light when probing exactly what it is that a user wants when he
or she comes asking for information. The reference interview process
is taught as a means of establishing the cognitive aspects of an
information request, but we can learn much more about the user and the
nature of his or her information needs if we ensure that we ask
questions about why the information is needed - and we may sometimes be
slightly surprised by the answers.
The fact that information needs exist is no guarantee that the
person who needs the information will take any action to find that
information. This is such an obvious statement that it seems,
sometimes, to take librarians by surprise! We are all so accustomed
to seeking and using information that we find the idea of not looking
for information when we need it quite difficult to understand.
However, all kinds of barriers to action exist: from simple inertia on
the part of the individual to formal organizational barriers to the
free flow of information.
We must also recognize that information-seeking can be understood in
two senses: it is a continuous activity in a generic sense, in that we
make sense of the world around us by gathering information, but, for
specific purposes it is, for the typical organizational member, a
highly spasmodic, event-driven phenomenon. And, often, the driving
event is a crisis - either for the individual of for the organization.
We cannot assume, therefore, that people have well-developed formal
information- seeking skills. It is more likely that the process has to
be re-learnt on each occasion - particularly if those occasions are
Informal information behaviour
Whatever the field, much research demonstrates that informal
information sources are likely to be explored first by the person
needing information and that formal information sources, such as
libraries, come no higher than second or third on the list and, often,
a good way further down.
Partly, this is the result of the fact that organizations are
information- processing or communication systems and that personal
interaction occurs for all kinds of work-related purposes and is
therefore a natural course to take when seeking specific information.
Information can also be a medium of exchange relationships: if I give
you information when you need it today, you may give it to me when I
Partly, however, it is also due to the spasmodic nature of
information need and the probability in an organization that there is
someone known to the information searcher who can be relied upon to
advise on the answer - can not only direct a seeker to other sources
but also comment on the validity, reliability, authenticity etc., of
information from those sources - and, most importantly, which
information not to trust.
Seeking information from formal sources
When it comes to seeking information from formal sources, such as
libraries, a number of things follow from what we know about users'
information- seeking behaviour:
- the crisis-driven nature of much information need, together
with the interrupted nature of much organizational work, means that the
user will always need the information right now!
- the same two characteristics mean that the user will always
fail to allow enough time to locate the information;
- because information-seeking is an irregular, spasmodic activity,
the user's knowledge of information sources and his/her recollection of
how to use them will always be deficient;
- because information serves affective as well as cognitive needs,
the information seeker may be unwilling to reveal the real
reasons for wanting information.
These facts, as well some of the other points I have been making
have some fairly obvious consequences for the design and delivery of
Service design and delivery
I think we need to begin by establishing that everyone agrees that
service delivery is a design problem: services ought not to happen by
chance, or be put together in a haphazard fashion, they ought to be
planned and, specifically, they ought to be designed around the needs
of the information user and his/her information-seeking behaviour.
We also need to be aware that re-design is necessary from time to
time (or, perhaps, given the pace of technological change,
continuously), since the organization, its members and its functions
change over time.
How, then, can an understanding of the information-user help us to
design services? First, we have to understand the nature of the
organization we work in and how its functions, aims and goals affect
the information needs of the people who work there - in other words, we
can begin by identifying organizational information needs. As I have
suggested, we can do this by looking at the nature of the value chain
for the organization and, thereby, identify the divisions or sections
that are most information intensive. We will find, of course, in some
organizations, that the information needed for a critical function, is
the organization's own information - internal information on, for
example, production, sales, or other issues. Whether we believe, as
information workers, that we have a role in managing such information
is an issue, of course, but - if no one else is doing the job
effectively - what is to prevent us from drawing the attention of
senior management to the problem and seeking to do something about it?
Having obtained some idea of the needs of the organization, we next
need to look at the nature of organizational work and assess the impact
of the organization on the information user. Again, as I have
suggested, we have a number of things to guide us: we know, for
example, that at managerial levels (and remember that management starts
low down the structure in many organizations) work is fragmented,
subject to frequent interruptions, and that the need for information is
often driven by crises of one kind or another. We also know that
individuals differ in their propensity to seek information and in their
knowledge of how to look for appropriate sources, and, when those
sources have been found, how to locate the needed information.
At this personal level, therefore, the information service must be
designed with these characteristics in mind. For example: airlines
have Frequent Flyer programmes - what about a Frequent User programme
for people who know what they want, how to get it and manifest a need
for information frequently - a fast access capability equivalent to the
quick check in; perhaps, in fully networked systems, personal passwords
to external information sources. Or, perhaps, an Infrequent User plan
would be more to the point - more personal service for the less
well-informed, a personal contact in the library or information
service, rather than the anonymous "Information Desk" - which perhaps
could be renamed the "Help Desk", in any event.
Thinking of Help Desk's lead's one to consider the application of
the computer software or hardware Help Desk - the phone number for the
terminally uninformed! Of course, these measures assume that you are
able to stratify your users into beginners and advanced classes, but
this is what finding out about users' needs and information-seeking
behaviour is all about. There is no point in doing it simply out of
casual interest - it has to lead somewhere and where it should lead is
to the redesign of services around the behavioural characteristics of
the users. Those characteristics differ, but they do cluster and
services can be designed for categories of persons.
Consider, as another example, the design of information products for
your users. Research we carried out in Sheffield some years ago
demonstrated that managers and other "desk-bound" people had certain
work characteristics in common: for example, most of their
"communication events" lasted only three minutes or less before
interruption, another task, a visitor, or whatever led to a change of
focus. This means that, in many organizations, you have three minutes
in which to get your message across through an information publication,
such as an information bulletin, for example (on-screen or printed).
The design lesson is that you design for brevity - keep it short and
keep it simple. E-mail messages with a screenful of new accessions
every few days will be better than a monthly printed bulletin.
We also found that meetings were a frequent focus for information
exchange and decision making and that the proportion of time spent in
meetings increased as one went up the managerial ladder, so that some
people could spent two-thirds of their working weeks in meetings.
Perhaps, then, the meeting, the working party, or whatever, should be
the focus of information delivery, rather than the individual.
I have tried to show that understanding the information user is a)
possible, and b) highly relevant to the creation of effective libraries
and information services. It is possible because we already understand
so much about information needs and the behaviour people engage in to
satisfy those needs, and we also know a great deal about how people
work. It is also possible because, being in direct contact with the
user, and having the capability to interact to an even greater degree,
the librarian or information manager has in his/her own hands the
capacity to discover more.
It is relevant because creating effective information services
demands knowledge of these matters - if we fail to understand users'
needs and if we fail to understand the processes of satisfying those
needs, information services will be ad hoc, unsatisfactory and,
eventually, ignored by the information user - when that happens the
information worker had better start looking for a new job because the
old one is not going to last much longer!
This document is copyright Professor Tom Wilson, ©1995