Citation
Measuring the value added by technical documentation

Material Information

Title:
Measuring the value added by technical documentation
Creator:
Mead, James Leland
Publication Date:
Language:
English
Physical Description:
v, 76 leaves : illustrations ; 29 cm

Subjects

Subjects / Keywords:
Communication of technical information ( lcsh )
Documentation ( lcsh )
Value added ( lcsh )
Communication of technical information ( fast )
Documentation ( fast )
Value added ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 72-76).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Science, Technical Communication.
General Note:
Department of English
Statement of Responsibility:
by James Leland Mead.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
39698006 ( OCLC )
ocm39698006
Classification:
LD1190.L67 1997m .M43 ( lcc )

Full Text
MEASURING THE VALUE ADDED
BY TECHNICAL DOCUMENTATION
by
James Leland Mead
B.A., Yale University, 1976
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Technical Communication
1997
AL


This thesis for the Master of Science
degree by
James Leland Mead
has been approved
by
Suzanne Schneider
Date


Mead, James Lelahd (M.S., Technical Communication)
Measuring the Value Added by Technical Documentation
Thesis directed by Professor James Stratman
ABSTRACT
Like any other business activity, technical documentation must add
value to a companys product or service. To gauge the effectiveness of the
technical documentation activity, and to justify its continued existence, that
value must be measured. The present study is a theoretical review of
principles by which the value of technical documentation can be measured.
Documentation value is the relationship of the cost of
documentation to the total value returned. The cost of documentation can
be established through careful benchmarking and proper accounting. Value
can be returned in one or more of three fundamental ways: reducing
internal investment (in documentation or other areas of the company);
directly improving return on investment, through increased customer
satisfaction and sales; and reducing after-sales costs, such as costs of
support. While studies exist showing value added in all three categories,
the second category, which shows the direct translation of increased
documentation investment into increased sales, is the most important and
deserves the closest study. Further, increased research in this area will lead
to an improved understanding of information as the product itself, rather
than simply a supporting component of product development.
Documenters seeking to establish the value of their efforts must focus most
closely on effective management of the documentation activity, including
benchmarking and record-keeping. Only data obtained through disciplined
management will afford accurate value-assessment acceptable to corporate
management.
This abstract accurately represents the content of the candidates thesis. I recommend its
publication.
Signed
in
fames Stratman


CONTENTS
CHAPTER
1. THE VALUE ADDED BY TECHNICAL DOCUMENTATION......................1
Thesis Question..................................................1
The Need for Research.......................................2
The Need for a Theoretical Review..........................3
Usability Engineering: An Instructive Contrast...................6
The Value of Usability Engineering: How is it Measured?....7
The Cost/Benefit of Usability Engineering and of Technical
Documentation...............................................9
How this Study is Organized.....................................10
2. CALCULATING DOCUMENTATION COSTS...................................12
Establishing a Benchmark for Documentation Cost.................13
A Basic Approach to Benchmarking...........................13
A Parallel Case: Usability.................................16
Benchmarking and the Entrepreneurial Department.................18
A Note on Accounting: Credit Where Credit is Due................19
3. HOW IS VALUE ADDED?...............................................21
Categorizing the Research: A Framework..........................22
Category 1: Reducing Internal Company Costs for Communication and Other
Activities......................................................24
Increased Internal Efficiency and Reduced Costs............24
Enhanced Product Development Processes.....................30
Corporate Transition and the Communicators Role: Cross-
Departmental Liaison and Repository of Company Knowledge and
Culture....................................................31
A Note on Accounting: A Corporations Intangible Assets....34
Category 2: Improving Direct Return on Investment...............36
IV


Direct Return on InvestmentAn Example...................36
Documentation Quality and Customer Satisfaction..........37
Documentation as a Sales Tool.............................45
Category 3: Reducing After-sales Costs........................49
Documentation and Phone Support at Intel: A Clear Case of Value
Added.....................................................49
Document and SupportFurther Evidence of Value Added......51
Documentation and Legal Costs.............................54
4. TOWARD A NEW VIEW OF THE VALUE OF TECHNICAL
COMMUNICATION................................................... 57
Assessing Value: An Approach...................................63
Conclusion: How Do We Measure the Value Added by Technical
Communication?.................................................66
The Value-add Categories: Summary.........................66
Implications for Further Research.........................67
How Can the Individual or Department Measure Value Added?.68
BIBLIOGRAPHY.........................................................i....'72-
v


CHAPTER 1
THE VALUE ADDED BY TECHNICAL DOCUMENTATION
In a business setting, technical documentation is not an end in itself, but rather a
means to an end. A technical documentation department, like any activity in a business,
exists to add value to the companys product or service. A business is the means by which
raw materials, labor, and intellectual capital are turned into money, and each separate
activity in the business must in some way contribute to that process, or there is no reason to
carry on with the activity. A manufacturing component of a business turns raw material
into a product or a part of a product that can be sold; a research component turns
intellectual capital into a form that will eventually add value to the business; a personnel
component nurtures the labor that will add value, and mediates between how much resource
the labor consumes in wages and benefits and how much it produces in income.
The technical documentation department must also contribute to the value produced
by the business. As with any other activity, if technical documentation cannot be shown to
contribute to the bottom line, it has no reason for being, regardless of its efficiency or the
quality of the documents themselves.
Thesis Question
How can the value added by technical documentation to a companys product or
service be measured? This study is a theoretical review of work on value in technical
documentation. The study reviews the principles by which the value of technical
documentation is established, and proposes the use of three general categories for
measuring that value to an organization.
1


The Need for Research
Technical documentation lacks a coherent body of research showing the value
added by technical documentation activities, and how to measure that value. Few studies
have been made of value:
The literature contains some individual case studies of value, of varying quality. Many
are largely anecdotal, though some, such as Hackos (1995), Cover, Cooke, Hunt
(1995), and Scholz (1997), include rigorous research findings. An overview of
techniques for estimating value added has recently been published (Redish, 1995).
Indexes to technical documentation conference proceedings (such as the ITCC) yield a
few presentations of individual case studies, generally without accompanying citations.
Rainey and Kelly (1992) in their survey of doctoral dissertations on technical
communication to 1990 cite no studies on value added by technical documentation, and
a search of abstracts in Dissertation Abstracts International (DAI) yields no
dissertations written between 1990 and 1996 focusing on value added by technical
documentation (several dissertations mention value in passing, or as a sub-topic).
No standard textbook or handbook on the subject of determining value-add for
technical documentation is cited in any of this literature.
Technical documentation as a discipline and as a company activity is increasing in
importance as advances in technology introduce new and unfamiliar products to the
marketplace, and as users demand greater usability in the products they buy. Further,
downsizing and cost-cutting throughout American industry makes the need for technical
documentation to justify itself in value terms ever more pressing. The problem of value
underlies all technical documentation activities in industry. From the point of view of
workplace trends, there can be few more important areas of research than the value added
by technical documentation.
Equally important is the problem suggested by the Pinelli and Barclay (1992, page
527): Against nearly all traditional, more rigorous measures of professionalism, technical
communication may not be a profession. Technical documentation is struggling to
establish itself as a true profession, recognized as such by the businesses that depend on
technical communicators skills. The last several years, for example, have seen a number
of active, and at times emotional, debates about certification for technical writers, and
2


whether that would ensure maintenance of skill standards and increase recognition of
technical documentation as a profession. No more effective step could be taken to
encourage recognition of the technical documentation profession than to show, with a body
of high-quality and persuasive research, that technical communicators add value to a
business enterprise in a way that others (who may try their hand at writing) cannot. Little
such research presently exists.
The Need for a Theoretical Review
As shown in the list above, some work on value added by technical documentation
does exist. Except for the Redish article (1995), however, there is nothing to tie together
the disparate approaches employed by authors studying their own firms. Value
measurement techniques and conventions have not been standardized; authors construct
techniques on an ad hoc basis for the work at hand, rather than relying on proven methods.
(The Redish article cited above is itself only in part a theoretical review; though the article
mentions some recent work on value added by technical documentation, much of the article
is a how-to description, supported by the authors experience and in part by previous
research.)
A theoretical review, as it is generally understood by social science researchers,
presents the theories offered to explain a particular phenomenon and [compares] them
with regard to breadth, internal consistency, and the nature of their predictions (Cooper
1989, page 12). Cooper goes on to explain that
...theoretical reviews will typically contain descriptions of critical
experiments already conducted or suggested, assessments of which theory
is most powerful and consistent with known relations, and sometimes
reformulations or integrations of abstract notions from different theories.
Cooper, page 12
3


A theoretical review of work on the value added by technical documentation must
track down all published research, and, more difficult, as much unpublished work as
possible, and attempt to synthesize a theoretical basis for making value measurements.
Another problem is the diversity of industries served by the technical
documentation profession. Will conclusions about value added for a computer firm by
software documentation be valid for a medical products firm? Will data about the value of
effective legal or insurance writing apply to the aerospace industry? One aim of research
must be to see technical documentation as a unified profession, and formulate theories that
apply to the profession as a whole, and not simply to a single segment such as computer
software documentation. An analogy is the well-established profession of accountancy. An
accountant is an accountant: he or she may have a sub-specialty in oil and gas, or in hospital
management, but the same general principles apply to each. The accountant uses
specialized techniques to add value to any industry. In the same way, a technical
communicator can learn a sub-specialty, but the value added in any case should be
measurable using essentially the same principles. To be effective, a theoretical review must
apply a theoretical synthesis across the technical documentation sub-specialties and show
how documentation value can be measured in more than one industry; it must not be simply
an ad hoc formulation applicable to one case or closely related set of cases.
A theoretical review and the need for primary research. Because few rigorous
primary studies have been carried out investigating the value added by technical
documentation, it is difficult at this point to establish theoretical principles. Thus, the need
for theoretical review and the need for primary research are circular: a theoretical review
will not be fully effective until more primary research has been carried out; and further
primary research can be stimulated by a theoretical review showing the holes in present
knowledge and suggesting possible standards of value measurement needing further
investigation. A theoretical review is needed now to assist in advancing this important
research. An integrative research review may also be needed at some time in the future to
synthesize the further work that will result.
4


Terms used in this review. This review addresses the value of product
documentation; documentation here refers to any textual or graphical material that instructs
the end user in how to use the product. This material can be presented in any of several
media: documentation until recently has most frequently taken the form of books, manuals,
reference cards, brochures, or other printed documents; documentation includes online help,
such as the Winhelp style of help familiar to Windows 95 application users; documentation
also includes recent innovations such as web-based information distribution. The common
characteristic of documentation for this study is that it is produced at least in part by a
writer, and often also by a graphic artist, and is for end user instruction.
Note that documentation generally does not include writing and production of the
product interface itself. For example, composing the terminology used for Windows drop-
down menus and other interface features is not considered a documentation activity.
However, the line between documentation and interface design is not always clear:
Interface terminology and layout decisions may be made by people who are trained as
writers and have documentation responsibilities. This is not surprising: documenters
have a sense of the end user and his or her needs, and are well-informed about what
cues the user needs to use the product.
With the rise of web-based interfaces, the lines separating product, interface, and
documentation are disappearing, as will be shown later in this study. For some
products, documentation and interface design are becoming the same activity. In this
study documentation is considered an activity independent of interface design.
Documentation tasks include research, writing, editing, graphic design and
production, document production, document testing, and other activities. As with
documentation itself, the distinctions among these activities are dissolving. Downsizing
and corporate cost-cutting have often reduced department sizes, forcing individuals to wear
many hats, and at the same time increasingly sophisticated technology, particularly
consumer publishing and graphics programs such as FrameMaker, Word, Photoshop, and
many others, has made formerly highly specialized tasks such as text layout and artwork
production accessible to many writers and artists. Still, some distinction can be made
between tasks such as writing and editing for the purposes of this study: writing, for
5


example, denotes the writing of any of the document types described above, while editing
refers to a subsequent cycle of correction or revision of the original text and layout. Other
tasks such as graphic design and document testing may be distinct activities, even if carried
out by the same person.
Usability Engineering: An Instructive Contrast
Accountancy was cited above as an analogy by contrast to the technical
documentation profession. A more useful comparison is with the closely related field of
product usability engineering.
Usability engineering and documentation may be seen as two forms of essentially
the same activity; documentation may even be seen as simply an aspect of usability
engineering and user interface design. Indeed, in some companies the two activities are
carried on in the same department, or by staff working closely together. Recent remarks
about usability engineering by one expert apply almost exactly to technical documentation
(Bias, 1990):
Especially in these competitive and economically constrained times, all
contributors to any sort of development effort must be able to justify their
contributions or be prepared to suffer the budgeting axe. This is
particularly true for human factors professionals, whose ultimate product,
usability, has a history of being considered a luxury.... The Usability
Testing Professional Users Group (now the Usability Professionals
Association) surveyed its membership and found that cost/benefit
analysis was the number-one concern.
Bias, page 4
However, unlike technical documentation, the usability field has recently inspired a
substantial body of work about cost/benefit analysis, and the value added by effective
usability engineering. It is interesting to note that, in the Redish article cited above about
value added by technical documentation, of 45 bibliographic references, some 15 are from
books and journals about usability, human factors, and human-computer interface design.
6


Further, an important text has recently appeared (Bias and Mayhew, 1994) integrating many
cost/benefit studies from the last two decades and describing standardized business methods
for calculating value added by usability engineering. The usability engineering profession
has had to overcome the same problems of recognition and acceptance that technical
documentation is struggling with. The work that has emerged on value added by usability
engineering activities may thus serve as a model for the work that must be carried on for
technical documentation.
The Value of Usability Engineering: How is it Measured?
For usability engineering, as for any company activity, value is the benefit of the
activity minus its cost. Usability engineering procedures are well established:
Methodological approaches to usability engineering have been evolving since the 1970s
(Mayhew and Mantei, 1994). Researchers including Gould and Lewis (1985), Mantei and
Teory (1988), Nielsen (1992), Shneiderman (1992), and Mayhew (1992) (all quoted and
summarized in Mayhew and Mantei, 1994), have developed well-accepted guidelines for
the usability process, i.e., scoping, functional specification, design, development, and
testing/implementation. Costs of these activities, particularly with regard to computer
software design, are known and measurable.
Usability benefits are more difficult to calculate than costs, but an accurate
approach can be made by dividing benefits into the following categories (Mayhew and
Mantei, 1994):
(For an internal development organization)
Increased user productivity
Decreased user errors
Decreased training costs
Savings gained from making changes earlier in design life cycle
Decreased user support
(For a vendor company)
Increased sales
7


Decreased customer support
Savings gained from making changes earlier in design life cycle
Reduced cost of providing training
Usability case studies have shown that calculations can be made for savings
realized in each category from effective usability engineering. A wide body of research
exists showing relative benefits of different usability features (for computer software, for
example, these might include use of color, choice of input devices, screen design, menu
structure, etc.); this research forms the basis for making relative and absolute benefits
calculations of usability design for a given product or set of products.
Using this cost/benefit framework, a powerful business case can be made for
usability engineering. A company can realize many potential benefits of usability
engineering (Karat, 1994):
Usability-engineered internal development projects can result in decreased
training, support, service, product documentation, and personnel costs, and
increased user satisfaction.... For organizations that develop and market
products for external customers, the financial benefits in the service areas
mentioned are realized in more indirect ways. For example, organizations
may be able to set more competitive service rates for customers while
realizing acceptable rates of return on investment.
Karat, page 50
Most important, however, is the accurate calculation of the actual cost savings
resulting from usability engineering. Sophisticated cost/benefit analysis techniques must be
used to convert the raw costs and benefits of usability engineering activities into a statement
of business benefit or cost over time, relative to other business costs. Calculation of
payback period, net present value (NPV) of inflows and outflows, internal rate of return
(IRR), and other techniques can show the actual benefit of usability engineering.
8


The Cost/Benefit of Usability Engineering and of Technical Documentation
Sophisticated cost/benefit analysis techniques are well-established for usability
engineering (many are described in Bias and Mayhew, 1994), and can be used to show the
value added by usability activities. How can these techniques show value added by
technical documentation activities?
Because usability engineering and technical documentation share many goals and
methods (technical documentation, especially online help, is often closely integrated with
the user interface, and can thus be thought of as one aspect of usability engineering), many
of the same cost/benefit calculation techniques can be applied to technical documentation.
For example, the list of usability benefits shown on page 7 also outlines a set of technical
documentation benefits. A body of primary research on the cost savings associated with
each benefit would suggest general guidelines for cost savings that can be applied across
companies, much as such guidelines have been formulated for usability cost savings.
Further, sophisticated accounting techniques must be used to model technical
documentation activities, in order to establish a common ground with other important
business activities. Most important, technical documentation professionals must learn to
speak the vocabulary of business, which is the vocabulary of cost and benefit calculation.
What Karat writes of usability (human factors) applies precisely to technical documentation
(Karat, 1994):
Human factors practitioners [usability engineers, technical documentation
professionals] must become proficient at communicating the idea that
successful use of human factors resource is an economic win-win
situation for the organizations they support, the end users of the usability-
engineered products, the customers who buy the products, and the human
factors practitioners themselves... In the past human factors practitioners
have often relied on descriptions of improvements in end-user performance
and satisfaction with usability-engineered products to communicate their
works value. Human factors engineers can increase their value to
organizations by providing financial data regarding human factors works
impact on organizational areas of concern.
Karat, page 46
9


Technical documentation professionals must learn the vocabulary of business (as
usability professionals are doing), in order to communicate to management the benefits of
their work in a way that shows that the work is relevant and worthwhile, and is Best
performed by trained professionals. An effective theoretical research review (as well as
further primary research), can help give technical documentation professionals both the
vocabulary and the data to show the benefits of their work.
How this Study is Organized
As noted, value can be defined as the benefit of an activity minus its cost. When
both benefit and cost of technical documentation can be expressed in dollar terms, a dollar
value can be calculated. When, as in many studies, the benefit side of that ratio is
expressed in less-concrete terms (for example, as customer satisfaction), the calculated
value is equally vague, though it may still have a role in focusing attention on
documentation quality. The task of the present study, then, is to examine each side of that
value ratio and describe ways that researchers and practitioners have devised to assign
dollar values to both documentation benefit and cost.
Section 2 of this study discusses how documentation cost can be calculated, and
how benchmarks are established. The section uses the important Cover, Cooke, and Hunt
article (1995) to illustrate how costs can be calculated, using careful record-keeping and
strict documentation process management. The section also looks at the parallel case of
usability testing and discusses cost calculation for a usability test program.
Cost calculation is not always easy, but it is comparatively straightforward, as costs
of time and materials can generally be determined in most organizations. Benefit
calculation is much more difficult. Section 3 of this study proposes three categories for
evaluating benefit (based on categories suggested by Karen Schriver, 1997): reduction of
internal company costs; improved direct return on documentation investment; and reduction
of after-sales costs such as product phone support. While each category has a body of
10


associated research, the third category, reduction of after-sales costs, has the largest number
of supporting studies and has produced very specific dollar amount measurements of
documentation value. For organizations looking for immediate, easily calculable results
from increased investment in professional documentation, this category is probably the
most fruitful, at least in the short term.
Section 4 of this study takes a longer view of the role and value of documentation,
focusing on the second benefit category, improved direct return on documentation
investment. Information development is moving away from its traditional role as peripheral
to the product itself and merely a supporting add-on. Information is increasingly seen as
itself a product, with its own value, direct and indirect. The section describes the
contention of one department manager (Charles Breuninger, DuPont) that to see
documentations value only in a negative way, i.e., simply as a cost reducer, is inherently
limiting-after all, costs can only be reduced so much-and helps keep documentation in its
old supporting role. Information is most properly seen as a product itself, with its own
costs and benefits like any other product, and research should focus on supporting this new
view of the profession.
The section concludes with speculation about implications of current value-added
research for documentation professionals and their departments, as well as for scholars and
researchers. The conclusion proposes consistent and effective departmental management as
the persistent characteristic of useful demonstrations of value added by documentation.
11


CHAPTER 2
CALCULATING DOCUMENTATION COSTS
No meaningful measure of value added can be made without effective calculation
of the costs of documentation. In the simplest terms, the difference between the cost of a
document, and the income or cost-reduction it provides, is the value added by the
document. The literature both on technical communication and on usability engineering
provides examples of how direct costs can be calculated.
In the case of both fields, the key is to establish a benchmark, a measure that can
accurately describe a projects cost and be used to reasonably estimate the cost of a similar
project. Establishing a cost may appear deceptively simple: add up the hours writers spend
on the book and multiply by some cost-per-hour figure from corporate accounting to obtain
the total cost. This process may in fact give a rough estimate of a documents cost, and is
certainly better than no estimate at all, but the figure it yields is of limited value because its
assumptions are so broad.
Close analysis of the documentation process yields a complex view of many
interlocking steps and activities, all of which add some cost to the end product. The
benchmarking exercise forces precise division of larger activities (e.g., research the book)
into smaller, defined units (e.g., interview the product subject matter expert, read the
product functional specification, draft a publications plan, and so forth). Each unit has a
cost; the sum is the total cost of the publication, a benchmark. A cost figure obtained
through careful benchmarking can be used with greater confidence than a rough estimate of
time and materials, and is important to establish before attempting to determine the value
added by the publication.
12


Establishing a Benchmark for Documentation Cost
Murphy (1992) discusses the concept of benchmarking documentation costs and its
importance to measurement of documentations value at the Digital Equipment
Corporation. He lists larger areas in the whole documentation process that should be
benchmarked, including (Murphy, page 724):
Cost to develop various information sets
Productivity using different tools or processes
Cost of maintaining and revising the mature products
Cost of documentation storage
Cost of documentation conversion
Average number of pages in manuals produced
He also lists examples of data that should be collected, including:
Pages per day output
Average manual size
Median time to complete manual
Cost per page
Quality assurance ratio (editing time expended per writing time)
Only after the gathering of benchmark information for the documentation process
can the value added by a given investment be measured. In addition, benchmarking allows
calculation of the trade-off between quality and cost, so that sound business decisions can
be made regarding documentation investment and return. As long as benchmark costs are
not known, documentation will continue to be regarded simply as an expense, rather than an
investment.
A Basic Approach to Benchmarking
A team at the firm Cadence Design Systems has published a detailed analysis of the
process they use to arrive at a meaningful cost figure for publication development (Cover,
Cooke, and Hunt, 1995). While the main point of their article is to show the cost savings of
13


quality documentation over the total document life cycle, their cost analysis provides an
illustration of the importance of careful project management, and can be used to model
other document cost calculations as well.
The Cadence team begins their calculation by estimating average fully loaded
salaries of all involved in document production. The fully loaded salary is the average
actual salary increased by 50%, to account for benefits, company facility and support costs,
and management overhead. Thus, for example, writers and system engineers at their firm
each average $48,000/year actual salary, equivalent to $72,000 fully loaded salary. This is
$34 per hour. Similarly, edit and production personnel average $44,000/year, or $66,000
fully loaded, for $31 per hour. The review team, comprising engineering leads, averages
$60,000, or $90,000 fully loaded, for $43 per hour.
Next, the Cadence team divides the documentation process into individual tasks and
sub-tasks, and calculates the cost of each according to the personnel involved and the time
spent. A 300-page manual requires the following main tasks:
Writer researches and plans
Write/edit first 100 pages
Write/edit next 100 pages
Write/edit final 100 pages
Sign-off and production
Each task is broken into its subtasks, such as actual writing, review, edit, and
correction. The study uses a table like the following:
14


!tem People (rid.) Duration {hrs). ; Rate; ,. ($) . r^vGpst^ ($) k
Writer researches and plans
Writer interviews, R&D provides info 1 80 34 $2720
Writer drafts pubs plan 1 40 34 $1360
Write/edit 1st 100 pages
Write pp. 1-100 1 250 34 $8806
Review individual chapters 3 5 43 $645
Milestone review 3 .5 43 $64
First edit cycle 2 16 31 $992
Writer incorporates comments 1 20 34 $680
...
Total cost $23,865
Necessary documentation investment cost, from Cover, Cooke, and Hunt, page 82
The 300-page Cadence Systems manual, then, required an investment of $23,865 to
produce. (This is undoubtedly low, as the authors point out; for the sake of their article
they eliminated certain preparatory and QA steps that would have added at least 10% to the
total cost.)
Having established a detailed and fully documented cost breakdown for their
manual, the authors are able to show the costs of corrections later in the documentation life
cycle. They can then compare that added cost with the costs of the up-front investment in
better quality, to show the value that quality documentation adds.
The Cadence calculation process is extensible to other environments as well.
Average salary information is relatively straightforward; the more difficult task is to
accurately identify the tasks and sub-tasks involved in the writing of any given document.
In another firm, for example, documents might require extensive graphics or terminal
screen prints-graphics development time can be included as one or more subtasks. Extra
edit cycles may be needed; greater research time or product familiarization on the part of
writers may be needed; steps may be added to publication production. All of these can be
broken into detailed subtasks, each with its own cost.
15


The need for effective documentation process management. The key to effective
cost calculation is effective management of the end-to-end documentation process. Hackos
(1994) emphasizes the fundamental need for effective management, and what it means for
the manager:
Managing the publications-development process essentially means keeping
track, knowing how much time you have expended on your project
compared with what you have to spend, identifying what percentage of
your project is complete at any point, and knowing if you have met your
quality goals. Publications project management means measuring the
progress of your team against your original estimates of schedule, budget,
and resources. It means knowing exactly how far you have come and how
far you need to go to reach your destination.
Hackos, page 21
Only with effective management can costs be calculated in a way that is reliable
and repeatable, and that can be used to show the value added by quality documentation.
A Parallel Case: Usability
Interestingly, usability practitioners use nearly the identical technique to estimate
the costs of usability activities. The tasks of a usability life cycle are defined, and then
divided into subtasks; the hours required for each task are estimated and multiplied by the
fully loaded cost of the appropriate employee; and the costs of all the subtasks are added to
estimate the total cost of the usability project.
A usability investment estimate might look like the following (Mayhew and
Mantei, 1994, page 19):
16


User surveys/questionnaires Development of survey: 40 hours @ $35/hour Pilot testing: 40 hours @ $35/hour Distribution and collection: 20 hours @ $15/hour Responding: 80 users for Vi hour @ $25/hour Coding and entering data: 20 hours @ $15/hour $1400 $1400 $300 $1000 $300
Usage study, simulation test, or prototype test Development of test: 40 hours @ $35/hour Pilot testing and revisions: 40 hours @ $35/hour $1400 $1400
Total cost $82,145
As with accurate documentation cost calculation, careful control and management
of the usability testing process is essential. Time estimates for individual tasks are not
standardized; only experience with the individual tasks as practiced in the environment at
hand allows accurate estimates. Accurate estimation of investment, in turn, is at the heart
of approaches to cost-justification for usability testing, as well as documentation.
Putting the numbers in context. With usability or with documentation, it is
important to use basic accounting techniques to put costs in proper context. A number such
as the usability testing cost cited above ($82,145) can seem very large, taken in the context
of a small departments daily expenditures. Costs for major documentation projects are
often far larger; for example, the cost for a single hypertext information product developed
by a team at Stewart Title Guaranty Company (Merritt 1997) was over $500,000. Such a
cost must not be viewed as a one-time expense, but as a capital cost, subject to net present
value (NPV) calculations just like other product development costs. When compared with
the NPV of a product expected to produce a revenue stream over several or many years,
even a half-million dollar expenditure can seem small relative to the total product revenues
expected. Successful use of NPV accounting methods is shown in the study of a recent
documentation project at DuPont (Knodel 1997):
I calculated the cost for developing, producing, and distributing the
technical bulletin and compared it to the NPV for the new chemistry
method. The cost of the information product was negligible compared to
the NPV; therefore, it was easy to convince the marketing manager to
proceed with the project.
Knodel, page 37
17


Benchmarking and the Entrepreneurial Department
Refinement of documentation benchmarking techniques, with an increasingly
accurate understanding of the true costs of quality documentation, can lead to an
understanding of documentation as investment rather than as ongoing expense. This view
of documentation can also be characterized as the services model, in which documentation
is a service to be purchased, and which provides a return; the opposing view is of
documentation is the support model, i.e., documentation as simple expense, like clerical
services or building maintenance.
An example of the services model implementation for a documentation department
is described by Edward See (1995). KBs division at IBM agreed to implement a services
model documentation department, in which the department acted as an autonomous
entrepreneurial provider of documentation services to other departments. Rather than
simply responding to another groups documentation need, Sees department developed
documentation proposals, complete with costs, and presented these to the department
seeking their services. That department, in turn, could purchase one of the proposals (or
none at all); the money for the purchase came out of one departments budget and into the
other, just as in an external purchase/sale agreement. For its part, the documentation
department then had to deliver, on time and within budget, the agreed documentation
services.
See describes the benefits of the services model of organization:
By developing an entrepreneurial model, organizations can develop a
quantifiable statement of the value of their services and products to the
larger businesses in which they function. Developing businesses within
businesses has become a significant way of understanding costs and
targeting internal investments.
See, page 421
The entrepreneurial model forces precise calculation of documentation costs, and
demands that documentation managers live up to their estimates.
18


See goes on to describe a more subtle advantage of such an arrangement within a
company: Support costs for a product are often seen as secondary costs, difficult to
quantify, and lost to actual product development. However, in a services model
documentation is a deliverable, like any other aspect of the product, and has costs and
returns clearly associated with it. Investment and return are very visible. See describes
how, in this case, the value-add of documentation becomes clear to developers, and how
...information development groups can develop a business strategy that
shows how best to invest in the information portion of products ... We
have also found that using a services model increased the organizational
autonomy and professional respect of the information development group.
See, page 423
Because it may require a considerable change in organizational processes and
culture, Sees approach is likely to be difficult to implement in most companies. However,
it provides a powerful incentive and model for proper calculation of documentation costs
and effective managerial control of the documentation process budget.
A Note on Accounting: Credit Where Credit is Due
Its important to note an accounting anomaly that can frustrate attempts to fix an
accurate unit cost for documentation (Redish, 1995):
.. .many accounting systems are still based on a manufacturing model, not a
labor-intensive service model. They use measures that relate overhead and
labor cost to the number of widgets that are produced or worked on. A
documentation group that is measured only on pages per day will appear to
be costing more if technical communicators are spending time on activities
other than writing and production or if their higher quality documents have
fewer pages.
Redish, page 37
A documentation manager may be likely to take quality into account in considering
the cost of a manual: while the cost for the manual might increase with added research
activities, for example, the resultant quality also increases; more was paid, but for a
19


different (and better) product. However, an older-style corporate accounting system,
Redish explains, may base all value on a measurable unit produced-in the case of
documentation, a page, for lack of any other clearly definable output. This may not take
quality into consideration (a page is a page...).
Thus, the relationship of investment cost to total return may not be accurately stated
in coiporate accounting systems, frustrating the efforts of a documentation manager to
show value added. Redish makes clear that, while there is likely little a documentation
department can do to affect corporate accounting practices, a documentation manager must
be aware of the appearance of documentation department activities on the corporate books.
She suggests that managers make a special effort to leam how their departments
investments are accounted for, and try to counter anomalies such as that described through
close communication with upper management.
20


CHAPTER 3
HOW IS VALUE ADDED?
With good management and careful record-keeping we can gather data about the
cost of documentation, as shown in Section 2. Measuring value added by documentation
means measuring the total value returned minus that cost. How do we measure total value
returned? From a research standpoint, the question becomes one of measuring the change
in value (the dependent variable) caused by a change in documentation quality (the
independent variable).
The independent variable is generally measurable: it is possible to isolate many
(though not all) aspects of quality in documentation and gauge their change. Quality of a
users guide might be improved, for example, by:
Increasing its factual accuracy (empirical analysis could determine the number of
factual errors in the manual).
Improving its readability with increased use of active voice, rhetorical devices such as
the given-new construction, and more suitable vocabulary level (all quality
improvements widely supported by research in readability).
Improving its use of graphics, white space, and layout and formatting (again, aspects of
quality generally accepted to improve document usability).
Changing the medium of presentation, as from hard-copy to online, and adding
hypertext references.
Adding supporting documents designed for different user situations, such as quick
reference cards.
Improving publication standards, such as paper, printing, and binding quality.
In addition to these research-based measures of improved quality in documents, a
recent questionnaire showed that documenters use the following five methods to measure
improvements in documentation (Ramey 1995, page 42):
21


Inspection for technical quality (formal validation procedures, technical review, QA)
Inspection for communication quality (editing, senior reviews, etc.)
Peer reviews
Usability testing
Field studies, contextual inquiry, etc.
The first three of these depend on expert judgment, and do not require research
outside the group or the company. More effective, and far less often used, are the last two,
which depend on data collection and analysis. It is these two methods-usability testing and
field studies of various sorts-that begin to approach the problem of measuring the
dependent variable, i.e., the value returned by improved quality, by measuring customer
response to the improved documents.
However, even if usability testing and field studies are used, the problem of
assigning a dollar value to documentation quality improvement remains. For most products
it is nearly impossible to isolate documentation as the single variable resulting in, for
example, increased sales or customers willingness to pay more for the product. Detailed
research studies showing direct value-add are rare.
Nevertheless, the literature contains studies showing a relationship between
documentation cost and value-add in measures such as reduction in phone support costs and
increased internal efficiency. Because such studies generally show only an indirect link,
and are often anecdotal and without hard data, they may not always be taken seriously by
corporate management, or even by documentation departments themselves. Still, such
studies can be used to suggest important documentation value-add research areas.
Categorizing the Research: A Framework
Redish suggests four categories to use for considering value-add studies: outcome
measures; customer satisfaction ratings; estimates of value added based on historical data;
and general perceptions of the value of documentation (Redish 1995, page 28). However,
22


only studies in the first category, outcome measures, can be used to measure the dollar
value of documentation improvement. Schriver suggests three useful categories into which
outcome-based value-add studies can be placed (Schriver 1997):
[Category 1] Reducing the investment spent on communication (e.g.,
more pages written per day for a lower cost, lower costs for printing)
[Category 2] Improving the companys return on the investment (e.g.,
increasing sales, productivity, or customer satisfaction)
[Category 3] Reducing the companys after sales costs (e.g., fewer
customer complaints and returns, lower training costs, fewer requests
for maintenance or calls to help lines, less litigation)
Schriver, page 77
Each of these activities is outcome-based; they can make a real, dollars-and-cents
contribution to company results, even if indirectly. Nearly all studies researched for this
paper contribute to the corporate bottom line in at least one-and often several-of these
ways. The following sections describe each category in detail, and cite studies that fit into
each.
These categories are also useful as a starting point for further research. Because
they are outcome-based, studies that accord with the categories may provide results useful
in business settings. More important, the three categories give to an existing-and future-
body of research a coherence and weight that may be missing from a collection of studies
whose relationships to each other are undefined.
23


Category 1: Reducing Internal Company Costs for Communication and Other Activities
Studies in this categoiy show savings of a variety of internal company costs
through enhanced documentation. Savings might be in communications processes (e.g.,
lower documentation department costs), or in increased internal efficiencies realized in
other areas of the organization, such as product development.
Documentation can reduce internal investment costs primarily in three areas:
Increased internal efficiency, in documentation departments and in other departments,
as well as reduced documentation production costs.
Enhanced product development processes, as with Rapid Application Development.
Easing of important company transition or upheaval such as downsizing, mergers, or
reorganization.
Increased Internal Efficiency and Reduced Costs
An important benchmark for studies of internal efficiency is the three-year study
carried out by JoAnn Hackos and Associates of the Federal Express Ground Operations
Policies and Procedures (P&P) manuals (Hackos 1995). In this project, researchers
examined in context Fed Ex employees use of the complex P&P manuals outlining all
operational procedures. The result of the study was an understanding of the high dollar cost
of the poorly written and organized P&P manuals. A recommendation for redesign and
rewrite of the manuals, accompanied by the cost figures obtained, was eagerly accepted by
company management.
The heart of the study was a usability test of the manuals with test subjects drawn
from the ranks of Fed Ex employees. Under controlled conditions, the test documented the
amount of time it took to find correct answers to questions using the P&P manuals,
employees unwillingness to use the P&P manuals because of their difficulty, and the
importance of task-oriented accompanying documents such as quick-reference cards. The
24


study also made use of interviews with many different subjects to establish how the
manuals were used in context and employees attitudes toward them. Finally, the study
included a heuristic evaluation of the manuals by usability experts to define their
shortcomings.
The usability test of the manuals with Fed Ex employees in which time to locate
answers was documented, combined with company pay and benefit data, allowed the
researchers to quantify the cost to the company of the inefficient P&P manuals:
On the basis of the initial usability test results, we calculated the potential
cost of lost time in searching for information. Based on a very
conservative estimate of two long searches per month and using the
average cost of an employee in each of the three job categories, we
calculated that it would cost the corporation at least $3,000,000 per month
in unproductive time. We did not calculate the cost of finding and acting
on the wrong answers to questions, but believe such cost to be high.
Hackos, page 325
Responding to these dramatic results, the company committed to a major revision
of the Ground Operations P&Ps, a large and expensive project. Interestingly, Fed Ex
rewrote the manuals using principles that would be familiar to any trained technical writer,
but that were apparently not considered important in the old edition. For example, they
organized the new manuals by task, rather than by policy; they targeted different sections
more precisely to the audience that would use the sections; they improved navigational and
access tools; they re-wrote text in second-person imperative and used active verbs; and they
improved graphic design and layout. In re-testing, the researchers found that the successful
search rate jumped from 53% of the time for the old manuals to 80% for the new manuals.
Further, with the old manuals only 38% of searches were completed successfully in under
three minutes, while with the new manuals 64% were completed successfully in under three
minutes. Using these figures, the researchers concluded that improvement in search times
alone within the primary audiences would save the company $400,000 in the first year in
productivity gains (Hackos, page 322).
25


For a very large, information-intensive operation such as Fed Ex ground operations,
the lesson is clear: up-ffont investment in professional technical communication and
company commitment to high-quality documentation can result in enormous value-add to
company operations. In this case, measuring that value also required a large up-front
investment of money and time (the study used professional consultants and took nearly
three years), but its results suggests both that the investment is worthwhile, and that gains
might also result from smaller studies in more limited areas of company operation.
Similar results were achieved on a large scale at Fisher Controls (Dunlap 1992).
While the Fed Ex improvements were exclusively intemal-the re-vamped documents were
used only by employees, not by customers-at Fisher Controls improvements were made to
customer documentation. Thus, the most important benefits were in improved customer
satisfaction. Still, Dunlap describes how a company commitment to improved
documentation processes resulted in huge savings even before the upgraded documents
reached the customers:
Fisher saved over $100,000, assured on-time delivery of documentation,
and supported extensive prerelease documentation testing (a first), which
permitted shipment of a major new product release three months early.
Early customers got documentation 95% faster than before.
Dunlap, page 700
Such savings were possible through, among other things, investment in upgraded
publishing tools and technology, which allowed writers improved access to up-to-the-
minute product information, and a shorter publication cycle. As with Fed Ex, the value
added at Fisher Controls required an up-front commitment by the company of time, money,
and expertise; the results were enormous savings over a relatively short period of time,
clearly showing the wisdom of the investment.
Investment up front in quality documentation can result in savings simply through
reduction in errors and in the need to issue corrections after publication. The Cadence
Systems study mentioned above (Cover, Cooke, Hunt 1995) quantified not only the initial
26


cost of writing and issuing documentation, but the cost of issuing Late Packs and Known
Problems and Solutions documents, as updates are known at that firm. Corrections issued
after publication were more expensive by a factor of nearly 10 times, their study showed,
clearly showing the value of investing in quality documentation from the beginning.
The Cadence study used several interesting techniques. In addition to the careful
calculation of the cost of writing a manual cited above, the researchers quantified the cost
of correcting defects in the manual discovered after writing but before publication: That
process consumed 540 hours of a writers time, at $34/hour, for a total cost of about
$18,000; an average of 150 defects were corrected in that time, resulting in an average cost
per defect of about $123. The researchers went on to calculate the cost of writing and
publishing a single major correction (along with the cost of phone support that resulted
from the original incorrect information), and compiled the following figures:
Item / Cost to correct
Cost per defect found m edit cycle $123
Cost per defect found in Beta $330
Cost per defect found in the field and fixed by a Late Pack $2,841
Cost per defect found in the field and fixed by a Known Problems and Solutions document $3,116
Included in the cost of corrections of documents in the field is the cost to the
customer of the original bad data, in loss of productivity and time spent by customer
engineers tracking down the problem. While the direct cost to Cadence was less than the
figures shown, the approach is a valid one: an engineering firm such as Cadence Systems
works closely with customers more as partners than as faceless market entities, and an
unnecessary cost to a customer through the companys error results in indirect cost to the
company, through loss of good will, through extra time spent solving the customers
problems, and through the less-tangible efforts that a problem in any close business
27


relationship entails. It is useful to regard waste of the customers time as the companys
problem as well, and to factor that waste into value-add calculation.
The literature contains many other reports of increased efficiency and reduced costs
through up-front investment in documentation. Edwards (1989) describes how a process
quality management initiative at AT&T reduced documentation costs by 53% and
document production time by 59%. As with other quality assurance approaches of recent
years, this effort introduced the supplier-customer model, a process description, and a
quality road map to management of the writing process. The relationship of these
techniques to those employed by others who have made radical improvements in
documentation value-add is clear; the supplier-customer model, for example, which is
simply a view of all internal and external entities for whom work is done as customers, is
carried to its logical extreme in Sees entrepreneurial department at IBM (described above).
The process description and quality road map, as described by Edwards, are simply
rigorous applications of consistent management, careful planning, and long-term record-
keeping, as described by Hackos (1994) and carried out by many others cited in the present
study. A similar approach is taken by Carbajal (1991): in her attempt to upgrade her
departmental operations at WordPerfect Corporation she concentrated on upgrading
writers tools. For her, however, the greatest value of improved tools is in firmer and more
consistent management of the writing process, including accurate record-keeping. It is
management and control that both adds value and allows that value to be demonstrated.
Some companies integrate writers directly into development teams, and this may
produce added value. Pieratti (1995) describes a study of three product-development
efforts, two using writers physically or organizationally separated from developers, and the
third using writers in close proximity to and contact with developers. Without question the
close integration of writers on the development team added value to the product:
The participants believe strongly that this interaction resulted in a better
product, increased ease-of-use, and a richer feature set. One developer
said, If there had been no involvement with the technical communicators,
[the product] would have been less functional and more obscure, would
28


have contained fewer features that were important to a product of this
nature, and would have been less usable by our customers.
Pieratti, page 66
While the Pieratti study does not contain quantitative data concerning the actual
dollar value of using the writer-developer teams, it suggests strong possibilities for genuine
savings, including particularly the ability to identify usability and interface problems early
in the development process and fix them at far lower cost, as well, of course, as a higher-
quality document set resulting from writers greater understanding of the product.
Reduced document production costs. In this context its important to mention
examples of reduced document production costs through up-front investment in improved
processes and hardware, and professional documentation management and skill. At Sabre
Computer Reservation Company, for example, Blackwell (1995) shows how task analysis
by a team of professional writers resulted in reduction of the number of pages in one
manual from 100 to 20-and the consequent savings in production costs of nearly $19,000,
more than paying for the effort that went into the task analysis. Note too the importance of
measuring writer productivity in units other than pages per unit of time-had such a measure
been used here, writer productivity would appear to have plunged. In fact, the rest of the
article makes clear, the shorter manual was a great improvement on the original, longer
manual, and resulted in improved customer acceptance of the product. A similar study by
Mazzatenta (1991) at General Motors showed the production cost savings in reduction of
one manual in length from 35 pages to two. The shorter manual was made possible by
professional writers analyzing anew an audience and a set of tasks that had been taken for
granted. Value can often be added in simple ways by professionals given the time and
resources they need and an enlightened management willing to realize that less is often
more.
29


Enhanced Product Development Processes
Writers roles in Rapid Application Development (RAD). Writers can have an
important team role in modem product development methodologies-they can, in fact, make
the difference between success and failure in several popular rapid software development
scenarios. McConnell (1996) describes a variety of rapid development strategies, including
evolutionary prototyping, in which a product is developed in close cooperation with major
customers, who are brought in not at the end of the process, as in traditional product
development, but from the beginning. The product is developed as a series of prototypes
shown to the customer for feedback and modification. The key role of documentation here
is in informing the customer on an ongoing basis about the evolving product, and, more
important, managing customer expectations of the product so that the customer understands
what is being developed and does not demand more than can be delivered. The writer is an
integral member of the team (as described by Pieratti), and acts as liaison between
developers and customers. Several other development techniques are based on the
customer-developer interaction of evolutionary prototyping, and in all of them the
documenter is a key player.
McConnell also mentions software component reuse, an important concept in
object-oriented program development. Component reuse can save development time and
money, but it requires close and careful management, and good documentation: One kind
of documentation thats especially valuable for reuse is a list of known limitations for each
component... .dont make people discover defects on their own when you know about them
already (McConnell, page 534).
The writer as tester. An important, and sometimes overlooked, role of writers in
product development is as Quality Assurance product testers. Most product development
organizations have Quality Assurance departments whose role is to test the product and
work with Development to eliminate flaws and bugs. Often, such departments are staffed
by highly paid engineers. However, by the nature of their work writers act as testers, even
if they have no connection with the formal Quality Assurance effort: writers must explore
30


each feature of the product in order to write complete documentation, and to do so they
generally use each feature of the product, most often in a pre-release version. Writers, then,
frequently discover problems in the product in the course of their regular work, and in that
way add considerable value acting as de facto testers. That this is a widespread
phenomenon is suggested in passing by Marchwinski (1997), in an article about a
questionnaire gauging attitudes to writers work and its value at the Attachmate
Corporation. Interestingly, while non-writers in the company overwhelmingly chose
Technical Support as the department to which writers added the most value (through
reducing phone support calls), the writers themselves said that they added the most value to
Quality Assurance activities at the company, suggesting that the company as a whole was
unaware of the importance of writers to product testing. That feeling is echoed in a letter to
the TECHWR-L mailing list:
Thorough documentation tends to be an effective part of the QA process,
and if it comes in early in the development cycle, the most efficient kind of
testing. I have always worn a testers hat when writing as a matter of
course, and no doubt I have saved my company plenty as a result.
Pete Kloppenburg, 28 April 1997
Corporate Transition and the Communicators Role: Cross-Departmental Liaison and
Repository of Company Knowledge and Culture
It is difficult to separate the writers important role in making rapid product
development possible from his or her role on teams and within the corporate structure and
context-both add value through enhancing the product-development process. The latter
role, however, as liaison and keeper of corporate knowledge, extends beyond development
of a specific product and into the smooth functioning of the company itself, in times of
stability or in times of difficult transition. As such, it is difficult to measure the value that is
added, but the writers role may be pivotal, and companies must leam to recognize its
importance and value.
31


Mirels study of in-house manual writers (1990) is of primary importance in
understanding the organizational importance of writers on a team. Written at a time of
massive corporate mergers and restructuring, as well as of introduction of desktop
computers to many business functions, her study shows the importance of context and
organizational behavior. As a professional analyst of context and the users needs and
assumptions, the technical writer can play a key role in maintaining smooth operation of
corporate processes and introducing change where necessary, particularly during corporate
mergers. Mirel shows how this happens through effective in-house communication; she
extends the point to show that writers can be liaisons among a variety of disparate
organizational units: Writers begin to incorporate themselves into the information
networks of the organization with their contextual analysis inquiries and their follow-up
communications to managers. As liaisons, writers and project managers can extend these
activities (Mirel, page 511). Having recognized the importance of their liaison role, the
article goes on to say, the writer should seek to strengthen and extend that role through a
variety of strategies, including matching strategies to circumstances, becoming members of
staff meetings, becoming innovation champions, and helping to construct support teams.
The benefit to the organization of a strong liaison role for writers is clear:
Effective project management and the appropriate design for a final
product depend on satisfying the needs of [many] diverse groups. To
achieve this end, writers must add a contextual dimension to their audience
analysis and act as liaisons in a variety of capacities. Unfortunately, these
activities often radically depart from the conventional roles that an
organization deems appropriate for a technical writer...Ideally, writers
pursuits will grow easier over time as an organization reaps the benefits of
more organizationally effective in-house documentation.
Mirel, page 524
A more general statement of the importance of cross-functional communication in
product development and smooth corporate function is made by Griffin and Hauser (1992).
Their study examined the progress of two automotive product-development processes, one
using traditional methods in which functions were isolated from each other, and the other
using a method that emphasized cross-functional communication. The authors conclude
32


that ...interfunctional communication is important to new product success (page 362),
and, more specifically, that
one explanation of the difficulties of achieving cross-functional integration
(Dougherty 1987) is that each function resides in its own thoughtworld-
engineers (R&D) speak a technical language of product features and
specifications and respond to an engineering culture of problem solving
while marketers speak in their own language.. .From both model and
research results, if projects are to succeed, then a product-development
process must bridge the thoughtworlds of engineering and marketing, and
promote freely flowing communication. Each function must understand the
needs of the other functions and provide the right information to meet those
needs.
Griffin and Hauser, page 362
The liaison role that Mirel suggests for writers is the bridge that Griffin and Hauser
see as essential for product-development success. Writers have a key position in the overall
function of the corporate organization, and must not allow themselves to be seen merely in
expendable support roles.
At the same time many companies were merging (the late 1980s and early 1990s),
others were downsizing, resulting in periods of profound change and upheaval for many
companies. While less dramatic, downsizing continues today throughout industry, and
documenters are often a target: Corporations are reducing technical communications staff
because they view the group or department as a net cost (Barchilon 1991, page ET-97).
Smudde (1993), Barchilon (1991), and others show the shortsightedness of this trend.
Documenters are a repository of knowledge in a company, and cannot be cut without the
potential for severe damage to the companys internal function and its image in the
marketplace: Departments entrusted with developing, disseminating, and archiving
technical data are central to company product-information activities and are the voice of its
knowledge to other people. Thus, technical communicators hold a position of
communicative power in companies (Smudde, page 36). Knowledge in our economy is
increasing in importance, and may even be viewed as a component of the product itself; to
33


cut technical documentation staff on the assumption that documentation is simply support
expense is to cripple the companys primary operation.
A more limited view of the disadvantages of reducing the documentation staff is
outlined in Steve and Bigelow (1993). The documentation function probably must
continue, in some way, regardless of restructuring -while it can be made more efficient, it
cannot be entirely eliminated. Thus, someone will write and edit, whether or not the writers
and editors remain at the company:
. ..the writer/editor is adding to the companys bottom line (i.e., increasing
profits) by decreasing the burden on technical personnel and providing this
service more cost-effectively....Managers, researchers, technicians,
secretaries, and others are more valuable to the company when they are
performing the work for which they were hired and trained, rather than
when they are assuming writer/editor functions. If they are not more
valuable in their intended function, you may wish to point out that these
individuals could be the focus of a future downsizing effort.
Steve and Bigelow, page 25
While it seems obvious that workers most efficiently perform the tasks for which
they are trained, this practical view is often missed by managers who fail to understand that
documentation is a function requiring time and professional skill; it is not something an
engineer can do in his or her spare time. Barchilon (1991) approaches the same problem
from a different angle: as a teacher of writing she finds herself in the difficult position of
teaching engineers how to write, in order to make up for management decisions to lay off
writers, and she sees clearly through her contact with engineers the continued importance of
professional communication even though the communicators themselves may be considered
expendable.
A Note on Accounting: A Corporations Intangible Assets
While beyond the scope of this study, it may be helpful to consider the work of
Karl Sveiby on properly evaluating knowledge-based organizations. Sveiby (1997) shows
34


how workers in knowledge organizations create not goods, but rather internal structures
(organization) and external structures (relations of any sort with customers and the outside
world). The creation and maintenance of such structures, along with the overall
competence of personnel, form the intangible assets of such a corporation, and the
repository of most of its value. By seeing the company structure in these terms, it is
possible to recast the profit and loss statement in terms of value added per professional, or
per employee. In this way writers (and other professionals) are accounted for in their true
light not as support expense, but rather as producers of value for the company, and a dollar
value on their overall contribution can be assigned by Corporate Accounting.
35


Category 2: Improving Direct Return on Investment
Studies in Category 1, reducing internal company costs, show that good
documentation can help a companys bottom line in indirect ways, by reducing related costs
and facilitating product development and corporate functioning. But can documentation
produce a direct, measurable return on investment? Category 2, improving direct return on
investment, contains potentially the most useful studies of value-add, but they are in turn
the most difficult to carry out, as it is so difficult to isolate documentation as the variable on
which increased product sales depends.
A glib answer to the question of documentations direct value lies in the enormous
success of after-market books for software and other products, including the phenomenally
successful Dummies books. People are buying documentation, and some publishing
companies are growing rich on it. Another trend that indicates the true value of
documentation-and companies attempt to profit from that value-is the rise in importance
of publishing subsidiaries such as Microsoft Press, Novell Press, Borland Press, and
Official Netscape Books. These subsidiaries do not produce accompanying user
documentation; they publish after-market books that users can purchase to leam more about
software packages. Microsoft, for example, now includes only very basic printed user
documentation with their most popular products such as Word, documentation often more
suited to marketing than instruction. Detailed, useful information about these products is
available only in online help, and in thick $40 paperback books sold in huge quantities by
Microsoft subsidiary MS Press and other publishers.
Direct Return on InvestmentAn Example
A more relevant case might be in the small-but-growing number of companies that
sell their product documentation, rather than including it in the price of the product. This
36


evidence is anecdotal-statistics for such companies are rare-but a parallel case is .
interesting: Mary Wise, at a leading supply chain management software company (personal
correspondence, May 1997) reports that her company markets training materials to
accompany their software product. These materials are expensive, but they sell like
hotcakes; the materials earn substantial revenue for the company. The customers who pay
thousands of dollars for the product information understand the importance of that
information and put a direct dollar value on it. The very close kinship between training
material and technical documentation (the two are often indistinguishable) makes this an
important case for documenters seeking to measure the value they add. Wises firm has
made training material a primary component of the product, rather than simply a support
add-on, and customers, when compelled to view the training information as a primary
component, are willing to pay for it. The companys approach has helped force a direct
demonstration of the value added by technical information.
However, because customers expect documentation to be included with a product,
many companies do not find it practical to sell the information components of their
products. Direct return on investment in documentation more likely takes the form of
increases in customer satisfaction-and sales-that can be traced to improved documentation
quality. Technical documentation can help translate customer perception into revenue;
while measuring the value added in this way is difficult, many studies suggest it is a fruitful
line of inquiry.
Documentation Quality and Customer Satisfaction
A study carried out by Smart, Madrigal, and Seawright (1996) lays the research
foundation for studies of the relationship between documentation quality and customer
satisfaction. In a carefully designed telephone survey Smart et al. traced a strong statistical
correlation between customers perceptions of documentation quality and their perceptions
37


of product quality, and found that their perceptions of document quality are related to their
product quality perceptions:
Logistic regression was used to determine whether satisfaction with
documentation was able to predict the customers perception of the product
quality of the computer hardware... The results of this analysis revealed
that customer satisfaction with documentation has a significant effect on
customer perceptions of product quality.
Smart et al., page 159
While the statistical correlation shown in this study seems clear, further studies are
needed to repeat and expand these results. A direct correspondence between customer
satisfaction with documentation and perception of product quality would be of defining
importance to the study of documentation value-add.
The impact of documentation on customer perception shown in this study has
important implications for product development and marketing strategies. Smart et al.
begin their study by describing the common position of documentation as a secondary
product component, without direct value, and thus vulnerable to cutting during any
corporate downsizing. Their research leads them to conclude that this may be a serious
mistake:
The effect of documentation on customer perceptions of product quality
suggests that companies should carefully evaluate documentations role in
overall company performance and decision-making. Even though
documentation seldom serves as a direct revenue producer, organizations
must not minimize its value....cutting documentation may eliminate an
important component of customer satisfaction used by customers to
differentiate products. Providing information about and support for
technical products remains a critical component of customer satisfaction
and product success.
Smart et al., page 159
This study raises as many questions as it answers. Most important, it strongly
suggests the need for research that can confirm its conclusions, which, though dramatic,
may be limited to a particular context or type of document or product. The study also
38


suggests the need to define quality, as it applies to documentation: Will an attractive yet
dysfunctional manual lead to a higher perception of quality even though the manual is
difficult to use? (page 161). Clearly, Jurans fitness for use definition of quality is an
important starting point, but the questions raised by this study show the difficulty of
defining quality more precisely. What does quality, or fitness for use, entail?
Attractiveness of layout and production? Accuracy of information? Clear, professional
writing style?
The difficulty of correlating quality with customer satisfaction is shown in a recent
study of a technical bulletin at DuPont that also used a telephone survey (Knodel 1997).
The bulletin under study had been developed by DuPont documenters in cooperation with
marketing specialists to assist in product technical sales, and was of high quality. The
researcher designed a telephone survey to elicit information about the bulletins direct
effect on sales, in an attempt to show the value the bulletin had added. Though responses
were very positive, little data was collected showing the bulletins direct effect on sales:
Most of the respondents were reluctant to judge how effective the bulletin
was at directly influencing the purchase decisionthey said they just didnt
know or cited other factors, both technical and sales promotional, that they
thought had more influence on customers behavior. However, the three
respondents (of ten) who did answer the influence questions rated the
bulletins as average to slightly above average in persuading customers to
try the new chemistry method. While this may not be statistically
significant, it still suggests that the bulletins were adding value to the
overall product.
Knodel, page 36
Even though she did not obtain hard results, this researcher considered the project
worthwhile, in that it confirmed the positive impression made by the document and focused
attention on the contributions of documentation to overall product success. It is worth
noting that DuPont has a reputation for enlightened management regard for product
documentation as an integral part of the product; she knew that her results, even though
they did not include specific dollar figures, would be taken seriously by her management,
because they were gathered in a systematic, thoughtful, and professional manner, and
39


indicated attitudes toward documentation that would translate indirectly into positive
product image in the marketplace.
Defining documentation qualitya framework. In an earlier article, Smart,
Seawright, and DeTienne (1995) define a framework for thinking about documentation
quality, one that they hope can facilitate a growing ability to measure quality. Quality in
documentation, the researchers show, falls into one or more of six major categories:
Transcendent quality, a purely subjective perception on the part of the customer or
user, nearly impossible to measure, but a strong influence on product perception-the
goodness of a product.
Design-based quality, a fairly objective measure of a documents adherence to
design specifications and document quality standards such as readability formulas (now
rarely used) and other predictive criteria.
Product-based quality, an objective measure of document features and
characteristics such as usability, reliability, accessibility, and the presence of search features
such as an index, or, in an online document, hypertext links.
Customer-based quality, or user-based quality, focusing on user needs and fitness
for use of the document.
Value-based quality, a measure of product excellence and customer satisfaction to
cost.
Strategic quality, a synthesis and extension of the other definitions, which can be a
factor in the companys competitive success in the marketplace, and which can include the
departments internal organization as well as managerial commitment and direction.
Smart et al. argue for a holistic view of quality, one that recognizes each of the
separate categories but understands that each adds value to the product. Further research in
quality can use these categories to more precisely define quality and quantify it. The
model, Smart says,
40


...provides a framework for showing the different emphases of vaiying
definitions and how they interrelate. Such a model can facilitate better,
more accurate communication between companies and consumers and
between practitioners and researchers-ultimately resulting in greater
success in a competitive, global market.
Smart et al., page 479
Measuring document qualitycriterion reference and prediction methods. Another
important approach to defining quality is taken by Schriver (1993). She articulates the
quality problem for documented succinctly: now more than ever, [writers] are finding
they must argue for the quality of the work they do, for the quality of the people who do it,
and for the value they add to the organization itself... [and] to present data to validate their
arguments (page 243). Where Smart defines categories of quality itself, Schriver
describes general approaches to measuring quality.
Actual measurement of quality is difficult and far from precise, but it is necessary if
document designers are to quantify the value of professional writing and graphic design
skills. Schriver describes the two dominant approaches to quality measurement: direct, or
criterion-reference measures, and indirect, or prediction measures.
Criterion-reference measurement directly assesses the quality of a text by
measuring what happens when its intended audience reads and uses the text-the text is
judged against some criterion. An organization might set a goal for the document (the
criterion); for example, a certain accuracy goal, or audience performance on a usability test,
or a certain score in audience rating exercises. The quality of the document is judged on its
performance against the criterion. Criterion-reference measurements are very powerful-
properly made, they can show in a direct way the effect of a good document, and lead both
to its actual valuation by the company and to further improvement and refinement of
document writing and production methods. Criterion-reference measures are, however,
difficult and time-consuming to make-they require field testing, and experience with
research situations and usability testing. Nevertheless, their importance in value-add
research is clear:
41


Research shows that the specificity of the information derived from direct
reader feedback is more useful in anticipating audiences needs than the
generalities more traditional methods provide. And what writers leam from
experience in usability testing can help them revise texts that have not been
tested... Consequently, researchers are exploring new methods that are
modeled on the diagnostic information that testing offers.
Schriver, page 244
Prediction measurement relies on a mathematical instrument designed to predict
how people might read or perform if they used a document (page 244). One familiar such
instrument are the readability formulas of Flesch and others, which assign a readability
score to any given text. In theory, the score can serve as a prediction of the ease and
effectiveness with which some defined reader can use the text. Probably more accurate are
the quality metrics developed in recent years by researchers in usability. Two sets of
quality metrics have been described in recent literature:
Hosier et al. (1992) devised a wide-ranging set of criteria for quality measurement
through research in cognitive psychology and many other fields outside of technical
documentation. Their metric relies on a number of taxonomic categories with empirical
support from the literature, including, for example, use of headings, voice, syntax, sentence
complexity, analogy and metaphor, graphics, and many other textual features. The
researchers derived a method of relating these features to document quality as determined
through reader usability tests; they claim that their evaluation method using the derived
quality metrics can track document quality improvement over time, serve as a diagnostic
tool for improving the quality of individual documents, and provide statistical evidence for
the value of professional technical documentation.
Downey et al. (1992) also devised an evaluation method, the U-Metric, that can
be used by writers to measure the quality of their texts. Their method includes a detailed
questionnaire about the text, a software program to evaluate the data, and even a Users
Guide to guide writers in their evaluations.
42


These examples of quality metrics are similar to readability formulas in that they
correlate specific textual features in a one-to-one relationship with document quality, but
their improved accuracy is based on recognition of the far greater complexity of reading
and using texts than that modeled by the word-length and sentence-length statistics of the
readability formulas. Developers of metrics promise no absolute criteria for analyzing
quality, but they do assert that metrics are objective diagnostic tools for identifying which
documents are best of class or which ones improve significantly over time (Schriver
1993, page 245).
Schriver concludes her wide-ranging paper with the assertion that quality in
document design makes a difference, and that continued refinement of quality evaluation
methods is an important pursuit. She includes two pages of provocative examples
illustrating the value of quality in document design. Some samples:
In Australia, by rewriting one legal document, the Victoria Government
saved the equivalent of $400,000 a year in staff salaries.
Southern California Gas Company simplified its billing statement and is
saving an estimated $252,000 a year from reduced customer inquiries.
In England, the Department of Customs and Excise cut a 55% error rate to
3% by revising some of its lost-baggage forms used by airline passengers.
The Federal Communications Commission rewrote its regulations for
citizen band radios and was able to reassign five employees who had done
nothing but answer questions.
Quoted in Schriver, pages 250-251
Another quality standardcustomer satisfaction. Quality metrics and predictive
measures of document quality rely on conformance to defined standards of design and
usability for quality measurement, and process efficiency is a hallmark of quality for any
documentation group. Fredrickson (1992) adds another standard to the quality model:
customer service. She advocates the continual monitoring of customer satisfaction as a
formal criterion for measurement of quality. Quality does not depend only on a static list of
standards and methods; rather, a documenters or departments relationship with the
customer (internal or external), and regard for the customers needs, becomes one element
43


of the quality delivered. Fredrickson lists ways of measuring that relationship. Her insight
into quality suggests the approach of Ed See, who made his documentation department fully
entrepreneurial within the company: for both, the information is a product with a set of
internal and external customers. Documenters work is judged, as any product in any
marketplace, by response to customers needs. This view makes documentation quality and
its achievement a dynamic process, depending always on customers:
Its easy to think that you, the deliverer of the service, know what the
customer wants. We can talk about quality, cost, schedule, project
management, and numerous factors in our attempt to meet customer
expectations. But these items may not be what the customers is looking
for. We must keep asking. We must listen to what the customer says, and
we must act on that information.
Fredrickson, page 397
Coe (1991) extends the idea of customer satisfaction as a factor in documentation
quality. For her, the documenters value-add is in his or her important role in achieving the
kind of overall customer satisfaction with a product that will translate into marketplace
success. She writes:
Corporations expect todays technical communicators to have more than
good writing and editing skills. They expect us to carry our full share of
the customer satisfaction responsibility. By taking the existing technical
writing paradigm and adding customer satisfaction-oriented components
like task analysis, audience analysis, usability testing, and audience
feedback, we can ensure our ability to meet this new corporate demand.
Coe, page MG-58
To the traditional write-review-revise-ship documentation cycle she adds task
and audience analysis, usability testing, and audience feedback, emphasizing the
importance of fitting the document to the customer and making these tasks as integral to the
writers job as writing itself.
Cover, Cooke, and Hunt (1995), in the Cadence study cited above, also emphasize
the centrality of the customer in defining and measuring documentation quality. They show
the cost both to the customer and to the developer of poor-quality documents: to the
44


customer, the cost of poor quality is in the waste of time and effort inaccurate or unusable
documents incur; to the developer, it is in extra support time and the immense cost of
revising published documents. The greatest value-add of good quality documents, of
course, is in increased customer satisfaction and the sales that is likely to bring, both from
the customer and from others who hear about your performance. As Coe says: Recent
marketing research points out that the dissatisfied customer tells between nine and 16
friends about your product, while another 13% tells 20 or more people (page MG-55).
Documentation as a Sales Tool
How else can documentation increase sales of a product, directly or indirectly, and
thus add to the companys bottom line? There is little research, but anecdotal evidence
makes it clear that good documents are increasingly being taken into account in buying
decisions. Reviews of software in PC magazines, for example, now generally review
documentation as well, and add documentation quality scores into total product
recommendation scores. (Examples can be found in any major computer-related
publication, including Computerworld, PC Magazine, InfoWorld, Windows Magazine,
MacWorld, and many others.) Care must be taken with such reviews: many are based on
only cursory use of the documentation by reviewers who may themselves be engineers and
developers (rather than documentation professionals who are more familiar with what
makes a good manual); the result is often a documentation review based largely on, for
example, professional presentation and the presence of a few helpful navigational aids such
as a good index. However, the mere fact that documentation is taken seriously at all is an
important evolution in complete product development and marketing processes.
In addition, there is direct evidence that companies are taking documentation into
consideration in buying decisions. One writer summarizes his companys enlightened
software purchasing policy:
45


We were ... able to show management how much a review of interface
design and documentation conducted by our team would save them per
function point of code they bought if such a review were built into the
procurement process...Our managers response? Wed be foolish not to
take you up on it.
George Hayhoe, letter to TECHWR-L mailing list, 20 June 1997
The writer is saying that quality and completeness of a products documentation
should be taken into consideration in software buying decisions, as well as completeness of
the code itself. This company realizes that documentation and interface is an important
component of the product, just like functionality, quality of code, and support.
Further, the documentation may be able to make up for other product failings. One
manager writes, it may take good documentation just to offset other issues and result in an
overall neutral effect, but if that's what it takes to avoid a negative customer impression, the
investment is worthwhile (Joe Vaske, Honeywell, 1997, personal correspondence).
Documentation and the Crossing the Chasm phenomenon. A final, and very
important, way in which documentation can add directly to company profit is suggested by
high-tech marketing guru Geoffrey Moore in his popular book Crossing the Chasm (1991).
The chasm Moore refers to is the difficulty of moving a new high-tech product from
acceptance by a small band of innovators and early adopters to acceptance by the far
larger and more powerful group he calls the early majority. The innovators and early
adopters are people who are interested in technology itself, and can see the possibilities of a
new product and its features for future needs-they are the visionaries of promising new
technology. The early majority, by contrast, are the pragmatists who are less interested in
the technology itself than in how it can solve their companies immediate problems, and
whether it will continue to solve those problems reliably and cost-effectively over the
longer term. (Compare the famous joke that no one was ever fired for buying IBM; in
other words, buying a mediocre solution from a well-known leader like IBM is less
hazardous to the cautious executives career than buying an exciting, but unproven, solution
from a company no one has heard of.) The circular problem for the young company with
46


the new product is that the eariy-majority pragmatists wont take a chance on the new
technology until they see that other pragmatists have had success with it first.
The problem for the new technology, then, is to establish sufficient marketplace
credibility that the early majority will begin to buy and use the technology. Moore writes:
When pragmatists buy, they care about the company they are buying from,
the quality of the product they are buying, the infrastructure of supporting
products and system interfaces, and the reliability of the service they are
going to get. In other words, they are planning on living with this decision
personally for a long time to come.
Moore, page 43
Quality documentation has a place in establishing the kind of credibility the
pragmatist is looking for. Innovators and early adopters, who may be fascinated by
technology itself, expect the new product to show up with little more than a set of
engineers release notes and a readme file; they probably prefer to tinker themselves with
the technology rather than read a manual anyway. The pragmatist wants a product that has
the kind of service and support that good documentation represents. He or she may also be
impressed by the investment and commitment that well-produced documents and
professional packaging suggest. In other words, good documentation helps establish the
kind of credibility and marketplace commitment that the pragmatist seeks in a new product,
and helps reassure the buyer that the company is in business for the long haul, and will be
there to support and upgrade the new product.
The following anecdote provides a powerful illustration of this point:
Documentation may have helped our company land our biggest contract,
ever. Our VP of R&D was entertaining potential clients from a company
that owns 28 hospitals (we write software for hospitals). These people were
also visiting our competition. Our VP had to figure out how to show that
our program wasnt just vaporware, as our competitions is. Being the
champion of documentation that he is, he instantly thought of showing
them our hard copy and online help. He pointed out: you cant have
complete documentation on vaporware! And it was true. The competitions
47


documentation was either minimal or non-existent. We are now their
preferred provider, or some such title.
(The VP told this story at a meeting of all the R&D people, which really
improved our standing, which was already pretty high!)
David Castro, letter to TECHWR-L mailing list, 28 April 1997
This is a classic case of using documentation to help cross the chasm, to establish
credibility for a pragmatic audience worried about the viability of the product.
Documentation added enormous value in this example; while it will rarely be so clear-cut as
in this case, documenters must consider their contribution to their companies viability in
the high-tech marketplace when calculating their contribution to company income.
48


Category 3: Reducing After-sales Costs
Categories 1 and 2 show documentation value through its ability to improve
internal processes and contribute directly to company revenues. However, the most
visible-and easily quantifiable-way in which documentation adds value is through
reduction in after-sales costs such as phone support. Studies have shown a direct
correlation in cost reduction through improved documentation. Such reduction is a result
either of a reduced volume of calls to support lines, or improved access to information by
support engineers, and thus less time devoted to individual calls. Category 3 studies
measure these reduced after-sales costs.
Documentation and Phone Support at Intel: A Clear Case of Value Added
Scholzs study (1996) of phone support costs at Intel shows the direct cost benefits
of more-effective technical communication. The writers role is extended from simply
developing print documents to generating new ways of delivering needed information to
customers and support analysts, such as through web databases and problem-tracking.
Scholz describes the active integration of technical communicators into the support process:
Technical communicators work with the [support] analysts to document
problems as they are reported, track the status of the research on the
problem, and write up solutions as they are developed. Technical
communicators also take part in the weekly analyses comparing the
frequency of electronic access with phone call volume. They help identify
the new problems generating the largest volume of calls to write up and add
to the database...
Scholz, page 37
For Intel, this integration of writers into the support process saves money in two
ways: through reduced phone calls to support, and through reduced time spent by engineers
per call. Because documenters have made information available by a variety of electronic
49


means, such as web-based information databases, customers can easily access answers to
many of their questions without resorting to a phone call. At Intel the cost of delivering
information electronically, as through a web page, is one-tenth the cost of a traditional
support phone call. The goal at Intel is a 70:30 ratio of electronic information delivery to
phone support delivery. Assuming 100 calls per day, these ratios produce the following
daily cost savings, using Intels support cost figures:
100 phone support calls @ $20 $2000
Assuming 250 work days per year, this results in a yearly savings of $315,000.
This, it should be noted, is for a single product. Intel, of course, makes many products,
some of which generate a high volume of support inquiries. For only 10 products, Scholz
points out, the savings is $3,150,000 per year. Up-front investment in professional
technical communication services makes electronic information available; a savings of over
$3 million per year is the direct bottom-line result.
Electronic support information, made available by the technical communication
department, is also used by the phone support engineers themselves. The result is faster
information searches and more-complete answers based on improved information collection
and distribution. In fact, electronic access saves customer engineers nearly 50% of their
time on the telephone per call, reducing the cost of a single call to $12. Using the $ 12/call
figure in the table above, instead of the $20/call figure, reduces the phone support cost per
100 inquiries to $360, for a savings of $1500 per day, or 75%.
Less-tangible value-add at Intel through commitment to quality technical
communication includes 24-hour customer access to information, improved customer
satisfaction, and easy access to the most up-to-date information. The value added simply
vs
70 requests handled electronically @ $2
30 requests handled by phone @ $20
Total
$140
$600
$740
Net savings:
$2000 $740 = $1260 per day
50


through reduced after-sales costs at Intel is clear and quantifiable; writers at Intel can show
that they have paid for themselves many times over.
Document and SupportFurther Evidence of Value Added
At the time of Scholzs study Intel had a sophisticated call support system in place
with costs well-known and verifiable. Gabberts study (1996) centers on a much smaller
company with almost no support call metrics in place. His study gathers preliminary
support call metrics, examines the design and production of a completely redesigned
manual, and then re-examines the support call numbers and costs, to directly quantify the
savings of improved documentation.
A thirty-day study of phone support before their product documentation was
redesigned showed that each support call averaged about $30, and one call was received for
every 18 units shipped. The manual redesign then began, using professional artists for
graphics and layout and far higher production standards, including four-color printing.
Their goal was to entice people to use the manual (there was considerable evidence that
users often called for support without even trying to find answers in the existing manual),
and to make needed information quickly and clearly available using information templates,
improved layout and graphics, and improved audience and task analysis. Overcoming
management skepticism at the increased cost of the manuals, the technical documentation
department finally released their fully revamped document. Though results were
disappointing after only six months, Gabbert shows that within two years, time enough for
the new documentation to have fully penetrated their user base, support costs decreased
dramatically. The calls-to-units shipped ratio dropped from 1:18, the number before the
new documentation, to 1:26, representing a savings of 44%. These simple figures
persuaded company management of the efficacy of technical documentation efforts and the
importance of up-front investment in effective information development.
51


Many studies similar to Gabberts show the reduction in after-sales costs through
effective documentation. Blackwell (1995), for example, describes the redesign of a
product installation guide for a product developed by a computerized travel reservation
service. Confident that the product could be easily installed, the developers declared that
no printed documentation was necessary. However, the documenters carried out a task
analysis showing that the target users would have great difficulty accessing the online
information as planned, and thus received funding to put out a printed installation guide,
saving the company large costs in support for product installation. Task analysis also
showed that the planned printed User Guide could be reduced from 100 pages to 20 pages,
for a production savings of nearly $19,000. Finally, a combination of user survey and call-
tracking was used to show a reduction in phone support calls from travel agencies who
were using the new product and documentation of as much as 80%.
Savings of after-sales costs through better documentation is not confined to
software products. In a study of two Veterans Administration offices, Daniel (1995) found
clear savings in the rewrite of a single veterans benefit letter that went out to all recipients.
The letter in this study described the disability pension that was available to veterans; phone
support was available from the VA offices for recipients who did not understand their
pensions. The author participated in an audience analysis and rewrite of the letter, and
gathered before-and-after data about phone calls related to the letter. The old letter, which
had been written by benefits counselors but without the aid of communications
professionals, generated an average of 1.5 support calls per letter sent (veterans who dont
understand the letters often make multiple phone calls seeking help); the rewritten letter
generated .27 calls per letter, a dramatic drop that allowed counselors to be redeployed on
other tasks, and that clearly showed the value that investment in professional audience
analysis and communication added.
Daniel makes another interesting point: in the software industry, company
accountants are generally keenly aware of the cost of such support activities as customer
phone help. However, in other areas (as with the VA), managers may not realize the
52


expenditures they are making that could be reduced by effective technical communication.
Daniel writes:
...before this study, VA did not realize how much veterans were depending
on telephone help to understand VA letters. Technical communicators
planning a value-added project might do well to think broadly of the ways
that clients (customers, users) are drawing on the organizations resources
so that they make sure they measure all the ways in which they might be
adding value. It is also interesting to find that reducing the need for
telephone support is a relevant measure of value added in all types of
organizations, including government agencies, not only in computer
companies.
Daniel, page 75
A detailed study by Spencer and Yates (1995) compared five similar product user
populations, each using the same software. However, four of the five used similar,
professionally prepared documentation sets, but management of one of the five refused to
use much of the supplied documentation and removed parts of it they deemed unnecessary.
The researchers kept detailed logs of calls for all five populations over a five-month period,
and the results are dramatic: The four groups of users who had the documents prepared by
professional communicators showed a very consistent set of support call rates for the time
period; the users who had the drastically shortened documents called at a rate some 40
times higher. Documentation was the only major difference between the user groups; on
the basis of this research, company management, using an industry-average figure of $15
per support call, concluded that good documentation was saving this division over $1
million per year.
Documentation and support callsan unintended effect. Anecdotal evidence
shows one unintended effect of improved documentation on support call costs that
documenters should be aware of: good documentation can increase support call costs,
though for reasons that are positive for the company overall. A recent exchange on the
TECHWR-L mailing list suggests the reason. One writer remarked that his highly praised
new documents had resulted in a 30% increase in support calls. The other writer
responded,
53


Good documentation reduces routine and repetitive queries, usually
about rough edges in the install procedure or the product itself, or hidden
assumptions about the users and their work environment. So why would
better docs increase the support load? If the previous version was too hard
to install correctly, many users may have given up and not used the product
at all; hence no support calls. Or maybe they managed to install the
product but only used the bare minimum of features because it was so
inscrutable or difficult to use.
Letter from Stuart Bumfield, 2 July 1997
Writers who see an increase in support call expenses apparently connected with
improved documentation should thus search for alternative explanations such as this, and
should attempt to tie the increase to the corresponding increase in customer satisfaction
with the product that increased understanding and use of features might bring.
Properly crediting documentation with cost savings. While the benefits of
decreased product support costs seem clear to communicators themselves, it is worth
mentioning again the point made by Redish (1995) about problems caused by misguided
corporate accounting practices. She encourages communicators to be sure credit is given
where due, pointing out that reduced expenses in one department (support) may not be
credited to the efforts of another department (documentation) on company books:
One problem is that many accounting systems still track costs by
department, not by project. If customer support costs go down, the
customer support group looks good. The documentation group doesnt get
any credit for reducing support costs, even if good documentation
contributed substantially to the reduction.
Redish, page 36
Documentation and Legal Costs
While reduced or improved product support is the most common source of savings
through quality documentation investment, another source is increasing in importance and
its appearance in the literature: savings in legal costs. While sporadic, such savings can be
54


enormous. In her article on documentation quality, Schriver (1993) mentions the liability
issue:
A growing number of product liability cases on record indicate that poor-
quality documents are costing manufacturers hundreds of thousands of
dollars.... Helyar and Camey report more than 50 examples of
organizations that have been sued because of factors related to poor quality
documentation.
Schriver, page 249
It is possible that liability issues in software documentation get scant attention
because there is little danger of physical injury with these products. However, in
manufacturing the danger can be great. Schriver cites one notable case, which resulted in
...a settlement in 1986 in excess of $945,000 paid by John Deere &
Company to a plaintiff who suffered severe and permanent injuries when a
crawler loader he was operating slid on an embankment and rolled over on
top of him. Deere & Company had failed to include appropriate warnings
and instructions in the operators manual.
Schriver, page 249
An important article by Smith and Shirk (1996) outlines the liability issue as it
concerns information developers for any product. In parallel with the increasing
recognition of information as an integral part of a product, a disturbing legal trend rather
broadly called information liability emerged in the late 1980s (page 190). The liability
applies to all information accompanying the product, not just safety instructions: What is
not generally known by business and technical communicators is that, in addition to safety
information, the instructions and advertising materials accompanying products may be
legally considered essential parts of the actual products themselves (page 189). Thus,
companies may be held negligent for poor-quality information
55


...if the duty of reasonable care is breached, for example, by inadequate
user testing-by inadequate determination of how the communication may
be understood and misunderstood by users-or by inadequate care in
ensuring that a user will receive, attend to, and understand the written
communication as well as safely use the product, the writer and the
manufacturer may be held liable.
Smith and Shirk, page 191
This liability extends to instructional and other information whose adequacy
depends on professional preparation:
Instructions may be defective if they exhibit poor use of language; illogical
sequence; failure to use art, typography, or design to highlight crucial
instructions or warnings; and failure to instruct users to keep instructions.
These are all aspects of defective product documentation that can be readily
corrected by attentive, competent communicators who are aware of the
need to avoid the perils of defective communication.
Smith and Shirk, page 197
Cummings and Kolesar (1991) make a similar point in their research on product
liability tort laws and the technical writer. They write:
Along with writing the warnings, the tech writer composes the instructional
literature that guides the user in assembly and use of the product. The
clarity of these instructions is as important as the safety of the product
itself. According to Charles Smith, if directions and warnings are
inadequate, there is liability since there is failing to appreciate and avoid
danger.
Cummings and Kolesar, page ET-165
The work of these researchers, as well as Smith and Shirk, makes clear that
investing in professional standards in writing is more than convenience or window-
dressing-it is a matter of legal liability, and huge sums may be at stake. Any technical
communicator seeking to discover how value is added to an enterprise through quality
documentation must consider carefully the effect of product liability laws and potential
damage awards.
56


CHAPTER 4
TOWARD A NEW VIEW OF THE VALUE OF TECHNICAL COMMUNICATION
We have seen how technical communication adds value in an organization by
reducing internal company costs, improving return on investment, and reducing after-sales
costs. These important documentation functions can produce measurable value added and
will continue to provide fruitful research data for documentation professionals seeking to
show their bottom-line contribution to corporate function.
However, the role of information development in a time increasingly dominated by
electronic hardware and software products and interchange is changing; any documenter
interested in defining value-add contribution must consider the evolving nature of
information and its primacy in our economic life. The traditional view of documentation
and its well-known roles may prove limiting as the profession evolves and redefines itself
in coming years. In the traditional view, information supports the product, but is not an
integral component of the product. This is particularly evident in the first and third
categories of value-add described above (reducing internal costs for documentation and
other departments; reducing after-sales costs such as phone support). These two categories
are the easiest to measure and contain the most supporting research, but they may reinforce
the view of information as support only.
The second category, directly improving return on investment, though difficult to
measure, suggests the new view of information as integral to the product, and even of
information as the product itself. Information can directly produce revenue; this realization
is taking hold at many progressive companies, and is shaping the future of our profession.
This insight is expressed by Charlie Breuninger, Manager of Technical
Documentation at DuPont. His view is worth quoting at length:
57


My general take on it [value-added research] is that most of what's been
done and measured so far has to do with cost avoidance-how much money
tech comm has saved someone because of fewer calls to the help desk,
quicker information retrieval (see FedEx), etc. And those are good things
to measure, if in fact they apply to your situation and you have a way to
measure them. I think they are the most obvious place for us to start, but
I'm afraid there may be a perception developing that cost avoidance is
pretty much it in terms of how we add value.
The problem is that if cost avoidance/savings is all you're measuring, the
expectation will be that you'll be able to just keep saving and saving
forever. And after a while, there's no more cost to be avoided or saved.
What do you do then? Have you added all the value you can?
I don't think so. I think that we're going to have to demonstrate some kind
of positive value added (cutting costs is a pretty negative way to add value)
to be viable in the long run. And I think the potential is there, but I don't
see anyone doing it yet.
What I envision is measurement of value added through increased customer
satisfaction and retention, increased brand loyalty, improved sales and
market share, and revenue generated directly by the information products
we create. These are all positive ways to add value, and there's no limit to
how much value can be added. But they're harder to get to and to measure
than the cost-avoidance approach. They will require the businesses we
work for to realize that their main products are information and knowledge,
not software, appliances, chemicals, etc., and to redefine their business
models accordingly. I think one way we can add value now is to help them
understand this.
I also think we're going to put our own limit on the value we add if we
continue to put ourselves in the tech comm box. I think we've actually
done a very good job of creating a perception of technical communication
as a profession, and that perception is full of parameters and
misconceptions that will hold us down. I think we would do better to
pursue the value added by usable, useful information and professionals
who know how to create it than to drag out that old tech comm warhorse.
Charlie Breuninger, personal correspondence, 24 March 1997
Breuninger sees that, fundamentally, the information is the product, in nearly the
same sense as the software or other item the customer is purchasing. Experienced
marketers know that the customer is not buying a large feature set or enhanced functionality
as much as he or she is buying the solution to a problem. The focus, then, is on both the
item itself and on the information that accompanies the item and allows the buyer to use it
58


to solve that problem. Neither information nor item is adequate alone. (This point is
implicit in the Crossing the Chasm high-tech marketing problem, cited above: for the small
group of innovators and early adopters who buy the earliest version of a new technology,
the information is secondary-they are intrigued by the product for its own sake. But for the
real market, the early and late majorities, the products ability to solve a specific problem,
rather than its slick features, is the only point of any purchase.)
Thus, information has a direct value, one that is difficult and important to measure.
That few hard figures are contained in the section of the present study addressing the
second category of value-added measurement (directly improving return on investment)
shows the difficulty. Direct value is a function of customer satisfaction, which produces
repeat or increased sales and a higher price. Customer satisfaction, in turn, is a function of
product quality, which, for documentation, is the ability of the document to help solve the
customers problem, through its clarity, accuracy, and fitness for the task and audience.
How can we measure documentation quality as it influences customer response, then? As
described, Smart (1996) uses statistical analysis of the results of a telephone survey; Cover,
Cooke, and Hunt (1995) use a count of factual errors per document; others use anecdotal
evidence of customer satisfaction and repeat business. In few cases, however, is there
direct, verifiable evidence that quality documentation produces a measurable return that
would not otherwise be present.
Still, direct return on documentation investment-the information as the product-is
the most important statistic to measure, as Breuninger points out, and provides the most
important area for continued research. Not only will such research help produce value-add
data with the most impact, but it will reinforce the growing importance of information
development as product component, and product itself. A complaint often heard is that
technical documentation is not taken sufficiently seriously as a development function and as
a profession; whatever the merits of the complaint, the best solution lies in this kind of
value-add research.
59


Writers themselves are increasingly conscious of the role of information as product,
and are working to spread that realization throughout their companies. A recent vigorous
discussion on the TECHWR-L mailing list contains a number of revealing comments. The
following, for example, shows how the information accommodates the product to the users
needs, placing proper emphasis on the user and his or her problem, rather than the product
and its features:
When companies make products and sell them, they generally think that the
product is what theyre selling. Nope. The product is a symbol for all or
part of the solution to their problem, and the information associated with
the product is what actually sells the product and tells the user how to solve
the problem. My concern for years has been why companies spend so
much money on the product and so little attention to the information.
The information describes the problem so that the user (or purchaser) can
recognize it as something he either has experienced, or may have to
experience. The information sometimes also describes the consequences of
that problem. The information then also describes how the product (or
whatever form the solution comes in) will solve or help solve the problem.
Other information (and this is what tech writers have done for decades)
shows/teaches the user how to use the product to solve the problem.
All of the above is information, and is vital to the sale of a product. So
why not pay more attention to how it does the jobs information is supposed
to do?
Elna Tymes, 27 June 1997
An even stronger statement of the primacy of information and the customers needs
is made by the same author:
The problem is fundamentally realizing what a company is selling. My
companys approach is that we only want to work with companies who
understand that basically they are in the business of selling information
supported by things-not things supported by information. Most companies
still think they sell things. We dont have time for them, when there are
enough client companies out there who understand the value of
information. THESE are the companies who will survive the current
information revolution. The rest will either go the way of the dinosaurs or
have to undergo a rather uncomfortable shift in perspective.
Companies who understand that they sell information are ones who work
closely with customers, who pay attention to how products are used and
what business problems they solve. The customers of these companies buy
60


products and services based on how well a particular set of problems can
be solved-and its the information that tells them how to solve the
problems, not the things.
Elna Tymes, quoted 23 June 1997 in a letter to Jennifer Geaslen
Two contributors to the debate wrote about the importance of this view to the
improved standing of technical documentation as a profession:
To take our job beyond mere competence, we have to look at the
information components of our products, and at what it is, exactly, that our
customers want from those products, and consider what information they
need, beyond what they already know, to get that. We need to consider how
to get that information to them in the best way we can. (Hint: manuals and
help files may not be the answers.)
Mike Huber, 30 June 1997
...to improve the standing of technical communicators in whatever arena,
the importance of what we do needs to be seriously re-evaluated by
industry (as industry drives public opinion, as well). I think emphasizing
the primacy and importance of information as a commodity in and of
itself is one good way to do this.
JeffWiggin, 30 June 1997
The convergence of product and information is driven from another side by new
technologies, particularly the emergence of the HTML and Java programming environments
and the popularity of Web- and browser-based information display. HTML-based tools
such as Microsoft FrontPage and NetObjects Fusion are becoming as important a part of the
writers toolkit as desktop publishing software. As with the Web, the display of the
information is the application, and the line between application programmer and writer is
indistinguishable. The information, what technical writers produce, is the product.
This is well described in a fascinating recent exchange on the TECHWR-L mailing
list. (Right brackets in the following quotation indicate the opposing voice, whom the
writer is quoting and answering.)
.. .there are no tools currently available for techical writers to develop
online help in a browser-independent, Java development environment.
61


I understand and share your concern and think we part ways over another
issue. My contention is that Netscape/Microsofit and so forth have a very
strong business case never to build such tools... Web pages are docs,
applications, and online help all rolled up into one. Why should they build
a separate environment?
Im all for honing technical skills, including some fundamentals of
programming and Java scripting. However, your post suggests that
we technical communicators need to become programmers.
Almost. Im not saying that we need to become programmers so much as
saying that:
The position of Technical Writer needs to be redefined, whether or
not this involves programming.
Web documentation has already and will continue to cause a drastic
shift in how Technical Communication is defined.
I, personally, am not willing to go there. My value is in knowledge
transfer, not in building tools and programming.
Its simply not efficient for me to do that.
Bingo! This is nicely stated, and I think youve put your finger on a
fundamental difference between us. I do not see knowledge transfer and
tool building as separated. IMHO, Technical Communication occurs
because someone usually a TW built the tools necessary to knowledge
transfer.
My value as a hard-copy writer lies in knowing when and how to build
the objects (tables, graphs, and so on) that allow knowledge transfer to
occur.
My value as a Web writer is no different. I know when and how to
build the objects (hyperlinks, Javascripts, Netscape plug-ins, and so on)
that allow knowledge transfer to occur.
Tool building is tool building. Technical writing is all about adaptation
anyway, and Im willing and able to adapt to any tool set.
As the line between writing and programming further blurs,
I agree that technical communicators are challenged to redefine
our niche. That niche should not be programmer wannabees.
I agree but there are plenty of multimedia experts and engineers who are
already building technically oriented Web pages. At a guess, unless we do
reinvent ourselves, someone else will cheerfully do it for us. I prefer the
former option.
DavidBlyth, 29 May 1997
Ruth Glaser, 29 May 1997 (quoted by David Blyth)
62


Web documentation has already and will continue to cause a drastic shift in how
Technical Communication is defined the writer says we must reinvent ourselves, not as
programmers, but as information/product developers-that is where technology and new
modes of information delivery are inevitably leading, and where technical communicators
will add the most value in coming years.
Assessing Value: An Approach
How can a documentation professional approach the task of measuring the value
added by documentation efforts? Many approaches are suggested by the researchers
quoted here; this study shows that they can be arranged generally in three categories. To
review:
[Category 1] Identifying ways that investment in documentation reduces costs and
helps the organization operate more efficiently.
[Category 2] Identifying the quality improvements achieved through documentation
investment that directly add value by increasing customer satisfaction.
[Category 3] Measuring the expenditures saved after product sale through effective
technical documentation, such as through reduced telephone support demands.
Breuninger and many others would contend that the second category, measuring the
value added directly by documentation investment, is the most important area of
investigation.
Carliner (1997) describes a practical framework that can be used for measuring the
value added directly by increased documentation quality. He laments the lack of any single
hard, verifiable statistic that can show the relationship between documentation and value:
Although our understanding of quality and value is maturing, and we have
begun to identify characteristics and perceptions that we can assess, we still
lack a cohesive framework for considering issues of quality and value.
Furthermore, we lack a consistently used, widely accepted methodology for
assessing the quality and value of our work.
Carliner, part 1, page 2
63


He then goes on to propose a framework adapted from an approach devised for
trainers, that shows the value added by their work. The framework has four levels of
assessment (adapted from Carliner, part 4, page 1):
User satisfaction. What did the reader think about the document? This level
assesses how users feel about a given communication product. This might be indicated
through comment forms included with manuals, though collecting reactions is often
difficult, particularly with online information.
User performance. Did the user leam the material as needed for effective product
use? This level measures the extent to which users can perform the main tasks.
Client performance. This level assesses the extent to which the communication
product met the business objectives. This is where value is added: How did the
communication product help the customer company improve its business performance?
Did it help them contain expenses? Generate revenues? Comply with regulations? It is
here that the communication has a specific value for the customer company, and thus a
value for the developing company. Carliner outlines how to evaluate client performance:
Assessing client performance requires that we identify a business measure
when we first plan a communication product for a client. Once we identify
the measure and the client agrees to it, we must also follow the changes in
that measurement. We usually begin doing so before publishing the
communication product and continue doing so long after publication. We
measure changes over time, rather than at a single point in time, because
the purpose of most business goals is facilitating a change over time, such
as an increase in revenue or a decrease in expenses. Unless we track the
trend in these measurements for a period before publishing the
communication product, we cannot, with any credibility, assert that
introducing the communication product caused a change in business
performance.
Carliner, part 4, page 6
Client satisfaction. The organizational equivalent of user satisfaction, above. This
level assesses how clients feel about the information development process in general and
the experience of working with you in particular. It becomes, in effect, a gauge of the
likelihood of the clients wanting to work with you and buy your products again.
64


This four-level framework underlies a practical method of assessing the effect of
increased document quality on a customer population, and thus the value that such an
increase adds. Carliner describes in helpful detail collection, analysis, and presentation of
data gathered for these four levels.
However, like other researchers, Carliner does not identify a single statistic or
methodology that can be counted on to show value added by documentation. The value of
undertaking this exercise, he says, is in:
Perhaps finding a set of data that will be persuasive to management. As pointed out,
this is not assured.
More important, instilling sound business practices in the management of technical
documentation efforts. Documenters must set goals, establish benchmarks, keep
records, assess results, and document successful methods so they can be repeated and
extended across projects.
Communicating results, along with our professional approach, to management and
other segments of our organizations.
As this list suggests, it is, finally, effective and professional management, and using
the value- and data-oriented language of management, that demonstrates throughout the
organization the value provided by quality documentation.
65


Conclusion: How Do We Measure the Value Added by Technical Communication?
As shown, the value added by technical documentation falls into one or more of
three general categories: reducing internal investment; increasing direct return on
investment; and reducing after-sales costs. This conclusion briefly reviews those
categories, describes some implications for further research, and lists some actions that a
practitioner can take to measure value added by technical communication in his or her own
organization.
The Yalue-add Categories: Summary
Reducing internal investment. As a cross-disciplinary and often interdepartmental
team member, the writer may be in a unique position in a company to facilitate company
processes and reduce costs through improved communication. The Federal Express study
(Hackos 1995) is a dramatic illustration of this, in its identification of ways to reduce
worker inefficiency through better documentation. Other studies have shown how up-front
investment in professional documentation activities can reduce document production costs,
enhance product development speed and quality, and even smooth difficult corporate
transitions in times of downsizing and restructuring.
Increasing direct return on investment. Identifying direct return on investment is
the most difficult, and the most useful, way to show the value-add of documentation. The
key is to show a correlation between improved document quality (i.e., documentation that
more closely matches customer need), and improved sales, achieved through customer
loyalty, ability to charge premium prices, word-of-mouth marketing, or in some other way.
A body of research has identified methods of measuring documentation quality through
criterion-reference measurement and prediction measurement; further, research has
identified categories of document quality (transcendent, design-based, product-based,
customer-based, value-based, and strategic) that can help accurately describe and measure
66


quality improvements. However, direct correlation of increase in quality with direct return
on investment is difficult, and is shown most powerfully in anecdotal evidence such as that
cited by Schriver. It is also important to consider the importance of quality documentation
to overcoming the sales barriers described in Moores Crossing the Chasm.
Reducing after-sales costs. Of the three categories of value-add, documentations
important role in reducing after-sales costs has been most thoroughly researched. Scholzs
study of phone support costs at Intel (1996) can serve as a model: she was able to identify
large cost reductions clearly resulting from the presence of documentation professionals on
the support team. Intels commitment to quality professional documentation in this single
case saves the company over $3 million per year, many times their up-front investment in
documentation. Many other studies show savings in after-sales support costs as well
because support costs can often be accurately measured, before-and-after studies involving
documentation improvements can yield precise, meaningful valuations of support cost
reduction. It is also important to note the growing importance of documentation in
questions of product liability, and the huge potential costs of liability suits that can result
from inadequate product safety documentation.
Implications for Further Research
The first and third categories of value-add (internal cost reduction and reduced
after-sales costs) have attracted the most study and produced the most immediately useful
results. Further study in these areas can add to this body of existing data. However, the
most important area for study is the second category, direct return on investment. We are
only beginning to gather meaningful data tying document quality to sales results, or even
data defining quality itself as it applies to documentation. Researchers such as Smart and
Schriver have constructed frameworks for thinking about document quality; research is
needed to further test the usefulness of those frameworks and the translation of data about
quality into bottom-line results. Finally, supporting work is needed to further understand
67


how information can itself be a product. Companies such as DuPont and Federal Express
are working toward this understanding of information; research is needed to identify other
companies that understand information in this way, and to study the impact that view may
have on company profits.
How Can the Individual or Department Measure Value Added?
This study has assembled a group of existing studies of value-add both to show
how successful studies can be carried out and to suggest a framework for studying
documentation value and its measurement. What practical steps can documenters take to
measure the value their own activities add to their organization? As this study has shown,
there is probably no one reliable method applicable in every case-no silver bullet for
value measurement (though measure of after-sales cost reduction can yield especially
powerful results). However, the study has identified the two approaches that any
meaningful study must take:
Search for an outcome-based measure of documentation value, and identify
communication activities within one of the three value-add categories that add
quantifiable value to the organization.
Implement disciplined management practices both in canying out documentation tasks
and in studying the value-add results.
Some practical activities that can be used to measure value are described below.
Using the value-add categories. The variety of studies cited, particularly in Section
3, suggest a number of fruitful approaches for an individual or department trying to show
documentation value in concrete terms. Documenters wanting to measure the value added
in their own organizations might ask the following questions:
(Reducing internal investment)
Has an up-front investment in documentation reduced publication production costs? It
requires professional skill and planning to effectively reduce document size. For
example, Blackwell (1995) and Mazzatenta (1991) showed how their departments were
68


able to reduce document size through thorough task analyses and careful rewriting, and
each saved their companies thousands of dollars in printing costs.
Similarly, has investment in professional time and skill lead to more effective use of
publication tools and document media, as shown by Dunlap (1992) and other's? Has the
documentation staff been able to put electronic document production and management
tools in place? Have documents been transferred from print to online? The savings
here can be internal (in-house documentation is more easily accessible throughout the
organization) or external (customer documentation is produced faster and with less
need for printing and service bureau expense).
Has better internal documentation saved time in other company departments? The
Hackos study of Federal Express (1995) showed dramatic savings of staff time through
improved internal documentation, resulting in savings of millions of dollars.
Has greater up-front investment in documentation reduced the need to issue corrections
and updates later in the product development cycle? Correcting an error in the field
costs from 10 to 30 times what it costs to correct an error early in development, as
illustrated by the Cover, Cooke, Hunt study (1995).
Are documented effective members of development teams, as shown by Pieratti
(1995)? Writers in their user-advocate roles can bring important interface development
skills to a development project.
As suggested by McConnell (1996), do documented act as liaisons with customers
during rapid application development efforts? Writers are often able to straddle the
thoughtworlds of both developers and customers, in ways that marketing and sales
personnel cannot, and assist with effective communication between the two groups.
This can reduce development time and increase customer understanding of and
satisfaction with the final product.
Do writers act as de facto product testers? This is an underappreciated role of
documentation staff; they are early and active users of products they document, and
much anecdotal evidence shows that they give important unofficial quality assurance
feedback to developers, improving products and saving later testing time.
(Increasing direct return on investment)
Has better quality documentation increased customer satisfaction? This is difficult to
measure, but it is very important to try. Anecdotal accounts show that the customer
comment forms often included with user documentation are rarely used and generally
only for complaints. It may be possible to contact customers directly, by phone or in
person; mail surveys can also be used. In any case, survey questions should be
designed to elicit not just suggestions for improvement but the relationship between
customer perception of the documents and customer perception of the product itself.
Researchers such as Smart (1996) and Knodel (1997) have surveyed customers about
customer perception, but it is difficult to systematize results across companies.
69


Can quality documentation directly assist with sales of new products? Any documenter
measuring direct value-add should be familiar with Moores Crossing the Chasm
(1991), and the potential role of documentation in selling a product to a skeptical early
majority, i.e., buyers who want proof of a products viability before committing to it.
The anecdote quoted on page 49 (David Castro) illustrates this powerful value-add tool.
(Reducing after-sales costs)
Is there a correlation between improved documentation and reduced support calls?
Scholzs study of phone support at Intel (1996) shows dramatic results; many other
studies also show call reduction and cost savings attributable to better documentation.
Daniel (1995) shows how this applies outside the computer industry (the Veterans
Administration, in this case) as well. A documentation upgrade project may present a
good opportunity to study this type of value-add; documenters must work closely with
support personnel to record and track support calls carefully in relation to product and
documentation releases. (Note that it is also important to be sure that reductions in
support costs attributable to improved documentation be properly credited to the
documentation department, not to Product Support.)
Does good documentation reduce the time phone-support personnel spend on individual
calls, or increase the quality of their assistance to customers? Scholz also showed how
an efficient online information database system made up-to-the-minute information
about products easy to access and use during customer support calls, cutting down on
the time expended on individual calls and reducing the need to call customers back with
more information.
Does the product present potential liability costs, and can poor-quality documentation
increase that liability? Smith and Shirk (1996) and others give examples of recent
product liability awards resulting from inadequate warnings and poor product
documentation.
Understanding the importance of good management. The common thread in any
effective and meaningful value-add study is careful, systematic, consistent management and
record-keeping. It is a not-inaccurate generalization to say that, while company executive
management most respects clear and well-documented data, in the absence of such hard
data executives respect controlled, professional management, with processes that can be
recorded and repeated. As Hackos (1994) shows, document process maturity is vital to the
smooth functioning of documentation activities in any organization; establishing value
begins with establishing regular processes and record-keeping, so that meaningful data can
be collected.
70


In short, the key to successful measurement of documentation value-add lies in
professional, consistent management over a long term. We have seen this in the work of
See, Hackos, and other researchers concerned with effective documentation project
management. As documentation professionals we will leam to assess our value, and
communicate that value to others, only through disciplined management of our activities.
That is, after all, the respect we owe our profession, and the only way to achieve the respect
others owe to us.
71


BIBLIOGRAPHY
Barchilon, M.G., How economics and technology have impacted the engineers role as a
technical communicator.
In Proceedings of the 38th International Technical Communication Conference, pp. ET97-
99. Arlington, VA: Society of Technical Communication, 1991.
Bias, R.G., and DJ. Mayhew, Cost-Justifying Usability.
Academic Press, Boston, 1994.
Blackwell, C.A., A good installation guide increases user satisfaction and reduces support
costs.
Technical Communication, 42(1), February 1995, pp. 56-60.
Blicq, R.S., Assessing the value added by in-house technical communication courses.
In Proceedings of the 44th International Technical Communication Conference, pp. 275-
278. Arlington, VA: Society of Technical Communication, 1997.
Carbajal, S.S., Customizing your tools: How to reduce your writing and production time.
In Proceedings of the 38th International Technical Communication Conference, pp. RT127-
130. Arlington, VA: Society of Technical Communication, 1991.
Carliner, S., Demonstrating the effectiveness and value of technical communication
products and services: A four-level process.
1997. URL: http://www.fredcomm.com/articles/value/abstract.htm
Coe, M.A., Customer satisfaction: New expectations of the technical communicator.
In Proceedings of the 38*h International Technical Communication Conference, pp. MG55-
58. Arlington, VA: Society of Technical Communication, 1991.
Cooper, H.M., Integrating Research: A Guide for Literature Reviews. '
Sage Publications, 1989.
Cover, M., D. Cooke, M. Hunt, Estimating the cost of high-quality documentation.
Technical Communication, 42(1), February 1995, pp. 76-83.
Cummings, M., and K. Kolesar, Testing the product: Usability and liability.
In Proceedings of the 38*h International Technical Communication Conference, pp. ET163-
166. Arlington, VA: Society of Technical Communication, 1991.
72


Daniel, R., Revising letters to veterans.
Technical Communication 42(1), February 1995, pp. 69-75.
Downey, L, J.C. Redish, R. Thornburg, A.G. Elser, The U-Metric questionnaire: A tool
for assessing the usability of manuals.
In Proceedings of the 39*h International Technical Communication Conference, p. 449.
Arlington, VA: Society of Technical Communication, 1992.
Dunlap, J., Customer satisfactionthe documentation challenge.
In Proceedings of the 39*h International Technical Communication Conference, pp. 700-
704. Arlington, VA: Society of Technical Communication, 1992.
Edwards, A.W., A quality system for technical documentation.
In Proceedings of the 36^ International Technical Communication Conference, pp. MG-43
- MG-46. Arlington, VA: Society of Technical Communication, 1989.
Fredrickson, L., Quality in technical communication: A definition for the 1990s.
Technical Communication 39(3), August 1992, pp. 394-399.
Gabbert, J., Reducing support calls and costs at Watlow Controls.
Intercom, June/July 1996, pp. 34-36.
Griffin, A., and J.R. Hauser, Patterns of communication among marketing, engineering
and manufacturing-A comparison between two new product teams.
Management Science 38(3), March 1992, pp. 360-373.
Hackos, J., Managing Your Documentation Projects.
Wiley and Sons, 1994.
Hackos, J., Finding out what users need and giving it to them: A case-study at Federal
Express.
Technical Communication 42(2), May 1995, pp. 322-327.
Hackos, J., Establishing quality benchmarks for technical publications.
In Proceedings of the 39^ International Technical Communication Conference, pp. 684-
685. Arlington, VA: Society of Technical Communication, 1992.
Hosier, W.J., P.M. Rubens, R. Krull, C. Velotta, Basing document quality standards on
research.
In Proceedings of the 39*h International Technical Communication Conference, pp. 428-
431. Arlington, VA: Society of Technical Communication, 1992.
73


Knodel, E.L., Using a survey to assess the value added by a technical support bulletin.
Intercom, October 1997, pp. 36-37.
Knodel, E.L., S.M. Gallagher, J. Leetham, T. Marchwinski, A.E. Smith, E.A. Sopensky,
Measuring value added: Rationale and approaches.
In Proceedings of the 43r^ International Technical Communication Conference, pp. 243-
245. Arlington, VA: Society of Technical Communication, 1996.
Kreitzberg, C.B., The electronic document in the informated organization.
March, 1993. URL: http://ignatius.lattanze.loyola.edu/lattanze/research/wp0393.015.html
Marchwinski, T., Using questionnaires to measure value added.
Intercom, May 1997.
Mayhew, D.J., and M. Mantei, A basic framework for cost-justifying usability
engineering.
From Bias and Mayhew, Cost-Justifying Usability, 1994.
Mazzatenta, E.D., Cultivating more quality and less quantity at the General Motors
Research Laboratories.
In Proceedings of the 38^ International Technical Communication Conference, pp. ET68.
Arlington, VA: Society of Technical Communication, 1991.
McConnell, S., Rapid Development.
Microsoft Press, Redmond, Washington, 1996.
Merritt, L., Is quality worth it?
Intercom, June 1997, pp. 39-41.
Mirel, B., Expanding the activities of in-house manual writers: Strategies for complex
audiences and purposes.
Management Communication Quarterly 3(4), May 1990, pp. 496-526.
Moore, G.A., Crossing the Chasm.
Simon and Schuster, 1991.
Murphy, S., Researching benchmarks for a corporate documentation group.
In Proceedings of the 39*h International Technical Communication Conference, pp. 723-
724. Arlington, VA: Society of Technical Communication, 1992.
Noonan, John V., How to escape corporate Americas basement.
Training, December 1993, pp. 39-42.
74


Pieratti, D.D., How the process and organization can help or hinder adding value.
Technical Communication 42(1), February 1995, pp. 61-68.
Pinelli, T., and R. Barclay, Research in technical communication: Perspectives and
thoughts on the process.
Technical Communication, 39(4), November 1992, pp. 526-532.
Ramey, J., What technical communicators think about measuring value added: Report on
a questionnaire.
Technical Communication 42(1), February 1995, pp. 40-51.
Rainey, K., and R. Kelly, Doctoral research in technical communication, 1965 1990.
Technical Communication, 39(4), November 1992, pp. 552-570.
Redish, J., Adding value as a professional technical communicator.
Technical Communication 42(1), February 1995, pp. 26-39.
Redish, J., and J. Ramey, Measuring the value added by professional technical
communicators.
IEEE Transactions on Professional Communication 36(3), 1993.
Sakson, D., and J. Sakson, The dilemma of reorganization.
In Proceedings of the 38^ International Technical Communication Conference, pp. ET68.
Arlington, VA: Society of Technical Communication, 1991.
Scholz, J., Adding value by developing information for online customer support services.
Intercom, October 1996, pp. 36-43.
Schriver, K., Dynamics in Document Design.
Wiley and Sons, 1997.
Schriver, K., Quality in document design.
Technical Communication 40(2), May 1993, pp. 239-257.
See, E.J., Moving to an entrepreneurial model: Providing technical information services
within a large corporation.
Technical Communication 42(3), August 1995, pp. 421-425.
Smart, K.L., J.L. Madrigal, K.K. Seawright, Documentation value.
Intercom, July 1997, pp. 35-36.
75


Smart, K.L., J.L. Madrigal, K.K. Seawright, The effect of documentation on customer
perception of product quality.
IEEE Transactions on Professional Communication 39(3), September 1996.
Smart, K.L., Cost-justifying technical communication.
Intercom, December 1996, pp. 38-39.
Smart, K.L., K.K. Seawright, J.L. DeTienne, Defining quality in technical communication:
A holistic approach.
Technical Communication 42(3), August 1995, pp. 474-481.
Smith, C.O., Products Liability: Are You Vulnerable?
New Jersey: Prentice Hall Inc., 1981.
Smith, H.T., and H.N. Shirk, The perils of defective documentation.
Journal of Business and Technical Communication 10(2), April 1996, pp. 187-202.
Smudde, P. M., Downsizing technical communication staff: The risk to corporate
success.
Technical Communication 40(1), February 1993, pp. 35-41.
Spencer, C.J., and D.K. Yates, A good users guide means fewer support calls and lower
support costs.
Technical Communication 42(1), February 1995, pp. 52-55.
Steve, M., and T. Bigelow, Coping with downsizing as a writing and editing group.
Technical Communication 40(1), February 1993, pp. 20-25.
Sveiby, K.E., The invisible balance sheet.
March 1997. URL: http://www2.eis.net.au/~karlerik/InvisibleBalance.html
TECHWR-L mailing list for technical writers.
Subscription available at LISTSERV@LISTSERV.OKSTATE.EDU; archives available at
URL http://listserv.okstate.edu/archives/techwr-l.html
Wiley, A.L., The quest for quality: Do you have a feeling theres more?
Technical Communication 39(3), August 1992, pp. 462-463.
76