THE NEW SCIENCE OF RECORDS MANAGEMENT

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP74-00005R000100020003-9
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 12
• Document Creation Date: December 9, 2016
• Document Release Date: April 24, 2001
• Sequence Number: 3
• Publication Date: October 1, 1954
• Content Type: REPORT

Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
The New Science of Records
Management
By ROBERT A. SHIFF and ARTHUR I3AIlCAN
REPRINTED FROM
HARVARD BUSINESS REVIEW
SEPT.-OCT. 1954
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
From In This Issue
Robert A. Shff and
Arthur Barcan
Industry today is spending a
lot of money on paper work; one
expert has estimated that it costs
20 cents to "create" one piece of
business paper, and a penny a
year to maintain it. The National
Records Management Council, of
which Robert A. Shiff is Execu-
tive Director and Arthur Barcan
is Vice President, has been pio-
neering in developing answers to
this problem.
In The New Science of Records
Management Shiff and Barcan
set forth some of the controls
which both large and small com-
panies can use to reduce the ap-
palling waste in record making
and record keeping (the authors
estimate that 65 cents of every
dollar spent to house records is
money down the drain), improve
efficiency, and meet the need for
better business histories. Stand-
ards, costs, and legal requirements
are also discussed in detail.
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
Edited under the direction of the Faculty of
THE GRADUATE SCHOOL OF BUSINESS
ADMINISTRATION, HARVARD UNIVERSITY
BUSINESS REVIEW BOARD
DONALD K. DAVID, DEAN, Ex officio
STANLEY F, TEELE, ASSOCIATE DEAN, Ex officio
' HOWARD T, LEWIS, Chairman
KENNETH R, ANDREWP DONALD T. CLARK
ROBERT N. ANTHONY FRANKLIN E. FOLTS
GEORGE P. BAKER BERTRAND FOX
CHARLES A. BLISS RUSSELL H. HASSLER
NEIL H. BORDEN MYLES L. MACE
EDWARD C. BURSK ROBERT L. MASSON
J. KEITH BUTTERS MALCOLM P. MCNAIR
Associate Editor: JOHN F. CHAPMAN
Assistant Editor: DAVID W. EWING
Managing Editor: VIRGINIA B. FALES
Advertising Director: ROGER B. KENT
Business Manager: T. K. WORTHINGTON
Circulation Manager: AGNES M. JORDAN
Address all communications, including manu-
scripts, to Harvard Business Review, Soldiers
Field, Boston 63, Mass. Telephone: KIrkland
7-9800.
Reprints of all articles and features in this
issue, as well as most previous issues, are
available. Information on request from Busi=
ness Office at above address.
A Volume Index to the Review is published
annually with the November-December issue.
The contents are currently indexed in The
Industrial Arts Index and in the Bulletin of
the Public Affairs Information Service.
Harvard Business Review is published bimonthly
by the Graduate School of Business Administra-
tion, Harvard University. Nathan M. Pusey,
President; Paul C. Cabot, Treasurer; David W.
Bailey, Secretary.
$8.oo per year, $z.oo per copy - United
States & Possessions, Canada; $ro,oo per year,
$2.50 per copy -- Philippines, Pan-American
Postal Union; $115.00 per year, $3.75 per copy
- elsewhere.
Make checks payable to the Harvard Business
Review.
Entered as second-class matter at Boston,
Massachusetts, Additional office of entry at
Concord, New Hampshire.
Copyright, 11944, by the President and Fel-
lows of Harvard College.
HARVARD
Business
Review
September - October 1954 Vol. 32, No. 5
ARTICLES
Big Costs of Little Fringes
Austin M. Fisher and John F. Chapman
Company Presidents Look at Their Successors
J. Elliott Janney
The New Science of Records Management
54
Robert A. Shiff and Arthur Barcan
Human Relations in a Bank
63
Chris Argyris
White-Collar Unions are Different
George Strauss
Executive Development:
In-Company vs. University Programs
83
Melvin Anshen
IIospitals Need Business Know-Ilow
John J. Brennan, Jr.
Cost-of-Living Pension Plan
Geoffrey N. Calvert
Will Your Business Die NA7ith Your
Edward L. Schwartz
Breakeven Point Control for I ligher Profits 123
Fred V. Gardner
In This Issue: Articles and Authors
The Editors
From the Thoughtful Businessman: Letters 21
Review Readers
Looking Around: Statistical Decision Alaking
Irwin D. J. Bross
Thinking Ahead: East-West Trade
Harold J. Berman
The Harvard Business Review does not lissiiine responsibility for
the points of view or opinions of its contributors. It does accept
responsibility for giving them an opportunity to express such views
PRINTED IN U.S.A. I and opinions in its columns.
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
the New Science
of Records Management
Q A plant engineer would be pained if a nut were screwed on by
one operator on the line, then loosened and screwed on again
by two more operators. Yet this sort of duplication is common in
paper-work processing.
By Robert A. Skiff
and Arthur Barcan
Paper work has turned into one of the costli-
est activities of business. It has become a serious
drain on both budgets and efficiency. When
scientific controls for paper work are lacking -
as they are in most companies - an average of
65 cents of every dollar spent to house records
is money down the drain; in other words, record
keeping is only 35% efficient. Moreover, the
amount of paper snowballs into an unwieldy
mass that strains every storage facility. The in-
evitable result is that executives find it increas-
ingly difficult to get needed information from
the records hoard.
This situation has caught the modern busi-
nessman unprepared. His predecessor of 75
years ago could keep his records under his hat
or on his cuff. If a subordinate needed to know
how or why, he had only to ask the boss - in
many cases, the founder of the company. With
the expansion of industrial and government ac-
tivity and the introduction of office machines in
World War I and after, paper work grew enor-
mously in bulk, but most executives still regarded
it only as a necessary nuisance, the inevitable
concomitant of progress. Now, by speeding up
clerical operations to a fantastic degree, develop-
ments in electronic equipment are threatening
to inundate business under a flow of paper that
will make its previous accumulations of records
look like drops in the bucket. A realistic ap-
proach to the problem cannot be postponed
any longer.
Fortunately for the company that wants to do
something about this situation, a number of
pioneers in records management have been
quietly assembling an organized body of knowl-
edge about records. They have been observing,
doing research, testing their ideas, validating
their conclusions, and putting into working order
the "pilot" installations of new records systems.
They have demonstrated that industry can -ac-
complish results like the following:
Only 12 weeks after the Columbia Broadcasting
System put a controlled record-keeping program in
operation, its savings totaled $41,000. The 33
million pieces of paper previously crammed into
its offices and files had been cut almost in half;
40 tons of paper had been carted away and sold
as waste.
Shrinkage of the records accumulation was only
part of the story. CBS had, for the first time, a
planned program of continuing records control. It
had a current index pinpointing every record in
every location. It knew what records to keep, and
why, how long to keep them, and where. Its per-
sonnel could easily make maximum use of the in-
formation stored in the records. '
Suppose that a conscientious executive wanted
to do more than deplore his creaking records
system. Suppose that he wanted to duplicate in
his own company (and it could be a large firm or
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
Records Management 55
a small one) the achievement of CBS. Where After a preliminary survey the Council re-
should he begin? What should he do? How ported to management as follows:
could he do it? What results should he aim for?
We shall discuss the answers to these and other
questions in the pages that follow, drawing in
large part on the wide and unique experience
of the National Records Management Council,
with which the authors are associated.
NRMC is a nonprofit organization established
in 1948 with the aid of a grant from the Rocke-
feller Foundation. It has had the advantage of
working in cooperation with such organizations
as the Kellogg Foundation, the American His-
torical Association, the Economic History As-
sociation, the Business Historical Society, and
the Society of American Archivists as- well as
leading universities; also, its program has been
administered by a board of directors consist-
ing of businessmen, archivists, and historians.
NRMC's work has now progressed to a point
where we can draw accurate outlines of the new
and helpful science of records management.
Gains in Efficiency
The science of records management encom-
passes three areas:
1. Records "birth control," which prevents un-
necessary forms and reports from coming into be-
ing, and limits record making to what is essential
for effective operations.
2. Record-keeping controls, which destroy the
papers that have outlived their usefulness, and in-
sure easy reference to and location of required
records.
3. Record-processing controls, which stream-
line the system, cut costs, and improve the quality
of required paper work.
In each of these areas there has been a good
deal of research and experimentation leading
to objective standards and exact procedures.
Moreover, it has become possible to forecast ac-
curately just what results can be attained by
applying scientific controls.
International Textbook Case
To illustrate what can be done, let us take the
case of the International Textbook Company.
Last August, the company's management asked
the National Records Management Council two
questions: (i) Did the International Textbook
Company have a records problem? (2) If so,
what action was necessary?
"Our measurement of your company's records
keeping totaled 8,227 cubic feet of records, or the
equivalent of 5,140 standard-size file drawers oc-
cupying over 7,500 square feet of office and stor-
age space. This is an estimated 20,550,000 pieces
of paper."
The audit report went on to estimate the records
improvement index (percentage of records to be
destroyed or transferred to low-cost storage) at
6 5 % to 7 5 % ; and the space utilization ratio (ratio
of cubic feet of records to square feet of floor area
occupied by files) at i. i to i, or about one-third
of full efficiency.
Guided by these estimates, the audit concluded:
"Lack of a sound and clearly defined records re-
tention schedule has caused a backlog of records
in offices and storage areas, with resultant uncer-
tainty as to what records may be destroyed, trans-
ferred to storage or microfilmed. . . . The appli-
cation of scientific records controls will result in
these benefits:
"1. Release 20% of the office space in which
records are currently stored; also release many cur-
rent records storerooms.
"2. Eliminate the need for new filing cabinets
and restrict the number already in use,
"3. Increase the utilization of storage space for
records from the existing ratio of i. i cubic feet
of records per square foot to the standard of 31/2
cubic feet to each square foot."
As a result of the study, International Text-
book now has a controlled record-keeping pro-
gram. Let us compare the results that were
actually achieved with the predictions:
Prediction Result
65%-75% turnover 72.9% turnover (32.9%
destroyed, 40% trans-
ferred to records center)
zo% of office space to be 22% (2,826 square feet)
released of space released
Release of many storage Areas with a rental value
areas of $2,342 released
Release of equipment 145 file cabinets released
(plus portions of other
equipment)
Greater utilization of stor- 8,588 cubic feet of rec-
age space ords (potential capacity)
stored in 2,450 square
feet of space - a new
ratio of 3.5 to x as
compared to the old of
1.1 to I
Company Size Immaterial
The International Textbook Company is by
no means an isolated example of what can be
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
56 Harvard Business Review
gained from the application of scientific controls
to paper-work problems that have haunted in-
dustry and government since the turn of the
century. Case after case can be cited showing
similar improvements. There is no significant
difference in results between large companies
and small companies, between private enter-
prise and local, state, or federal government
agencies. The government does not have any
monopoly on unnecessary paper work, and a
small company can be as badly snarled in red
tape as a large organization.
Some representative organizations that have
adopted scientific records management with
NRMC help and the results obtained as meas-
ured by the records improvement index are
shown in EXHIBIT 1.
EXHIBIT I. GAINS FROM SCIENTIFIC RECORDS
MANAGEMENT AT TYPICAL INSTALLATIONS
Organization
Records
destroyed
Records
transferred to
new records
center
Total:
records
improvement
index
City of New York
(pilot installation)
47%
26%
73%
Columbia Broadcasting
System
46
21
67
Electric Storage Battery
Company
23
27
50
El Paso Natural Gas
Company
28
37
65
Hamilton Standard Divi-
sion (United Aircraft
Corporation)
20
Irving Subway Grating
Company
69
7
Reaction Motors Company
9
6g
Scott Paper Company
44
21
Professional Standards
In some quarters there is a tendency to depre-
cate records management as a science, to feel
that effective records control requires no more
than good intentions and common sense. To be
sure, common sense may dictate that a records
system be kept in order; good intentions may
motivate willing workers. But the gap between
intent and accomplishment cannot be bridged
without knowledge. There is no "handy home
guide" that a well-meaning executive can follow
to streamline his records and reduce costs. Only
professional standards, research, and tested ex-
perience can provide him with a solid base for
making his decisions and choosing the approach
which seems best suited to his company's or
agency's needs.
Dealing with Specialists
It is because records management has become
a science that management will ordinarily find
itself dealing with specialists in the course of
developing controls for paper work. One of the
most critical junctures in a program occurs when
the specialists come to management with their
proposals. Just what should management ex-
pect of them?
Goal and Timetable. In the first place, if a
proposed program is backed by research and ex-
perience, management should expect a "balance
sheet" of what the program will accomplish and
what it will cost. It needs intelligent answers
in advance to these questions:
C What are the specific objectives of the pro-
posed program?
C What kind of work does it involve?
C How long will it take?
Q How much will it cost?
C Who will do it?
C What tangible benefits will result?
If these questions cannot be answered fully
and satisfactorily, management would have rea-
sonable doubts about the reliability of the pro-
posed program.
Practical Demonstration. Secondly, manage-
ment should be certain that the core of the sug-
gested program is an actual installation, a going
records system approved and understood by the
people who will work with it. It is easy for an
outsider to come in and state that a records sys-
tem needs repairs, that it is confused, unwieldy,
difficult to use. Likewise, it is easy to write a re-
port to management recommending that paper
work be simplified and systems reorganized. It
is quite another thing actually to effect the lower
cost and higher quality that mark an efficient
records system.
Continuing Control. Finally, the proof of a
sound program is the extent to which an organi-
zation can carry on successfully after the special-
ists have departed.
Records management is not a one-shot opera-
tion, over and done with as soon as the system
has been overhauled and rehabilitated. The new
paper-work controls should insure continuing
efficiency, continuing cost reduction. If the
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
Records Management 57
beneficial effects of records reform last about as agement an idea of how modest the total outlay
long as a New Year's resolution, something is can be:
wrong. A healthy program includes not only
instructions on current operations but also
preparation and education of company personnel
so that they themselves can adjust the program
to future needs.
Quantity of Records
One of the key questions for management is:
How many records, how much paper in our files,
should we have? The answer is important not
only in determining the need for a program but
in appraising the proposals of the specialists.
Naturally, an organization's need for records
depends somewhat on individual circumstances,
and even within the same industry or field of
work there will be differences. However, cer-
tain "rule of thumb" ratios have been worked
out on the basis of past experience and research
which executives can use as professional stand-
ards subject to local variation (note that i cubic
foot of records is equivalent to about 2,000
pieces of paper) :
Industrial operations (tex- i cubic foot of records for
tile plant, aircraft fac- each employee on the
tory, etc.) payroll
Government agencies or 5 cubic feet of records for
companies subject in de- each employee on the
tail to government reg- payroll
ulations (public utility,
airline, etc.)
Specialized clerical opera- 10-15 cubic feet of rec-
tions (accounting, -pur- ords for each employee
chasing office, etc.) on the payroll
Operating and Installation Costs
What are the costs of keeping and maintain-
ing records under scientific controls? The sav-
ings over conventional systems are striking. As
a professional standard, a model records center
with prompt reference service need cost only
$0.75 to $0.90 per cubic foot for a year. (These
figures include costs of personnel, space, and
equipment.) Compare this with the typical cost
of maintaining records in the office, which is
$7.50 per cubic foot per year, and with the aver-
age storage costs of conventional company sys-
tems, which is $3.00 per cubic foot per year,
and the magnitude of savings becomes readily
apparent.
But what about the costs of establishing scien-
tific paper-work controls? The following figures,
representing the approach of NRAMC to the as-
signments which it undertakes, should give man-
i The audit or survey ordinarily requires two
or three days. There is a charge of $8o per person
per day, plus travel expenses. There is no addi-
tional fee. The total cost does not ordinarily run
above $240.
Q The installation of new controls (including
the education of company personnel) is done on a
fixed-fee basis. The fee averages about $5o a day
per specialist. The number of specialists assigned
to the job and the time required varies for obvious
reasons. In the case of one large manufacturing
company, the installation required the full-time
services of two specialists and the half-time serv-
ices of a third for six months. In the case of an-
other large company, the job required the full-time
services of two specialists for slightly less than
three months.
Quality Control
Virtually every clerical operation (typing, in-
voicing, punching cards, and so on) these days
is a double-cost setup : first, the piece of paper
is produced; secondly, it is checked. This not
only is very expensive, but does nothing to pre-
vent future errors.
One of the impressive features of scientific
records management is quality control. Records
can be maintained in better shape with less
inspection by selecting only that portion of
the clerical product which needs verification -
when it is needed, and to the minimum extent
needed. Experience has firmly established that
i oo % inspection does not result in i oo % qual-
ity. On the contrary, research indicates that
ioo% inspection frequently fails because of
fatigue, boredom, and distraction. Quality con-
trol results not only in greatly reducing the num-
ber of full-time inspectors while improving the
condition of the files, but also, in some cases, in
enabling effective control on a part-time basis.
Furthermore, it will soon be possible, as a re-
sult of research now in progress, to tell manage-
ment in a matter of days what the quality level
of its recordkeeping is and what it should be,
and what maintaining quality costs in compar-
ison to what it should cost.
Business History
Business history is a little-publicized-branch
of records management; yet it is' extremely im-
portant from the viewpoint of both executive and
historian. The historian obviously is concerned
with source material to be analyzed, interpreted,
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
58 Harvard Business Review
and woven into an objective account of the eco-
nomic and political scene. The executive, on the
other hand, is interested in the business record
as the memory of his organization. Good rec-
ords of a company's experience, of the achieve-
ments and the failures of its people, are a cor-
porate asset. They provide invaluable back-
ground material, explaining why and how deci-
sions were made, and they also serve as adminis-
trative guides and references.
The necessity for keeping vital records and
not tossing away every piece of paper in an ex-
cess of housekeeping zeal is well illustrated by
the embarrassing plight of one of the largest
banks in the world. Engaged in an antitrust suit
with the government several years ago, bank of-
ficials were forced to go, hat-in-hand, to the
government and ask for copies of their own rec-
ords to prepare their defense. The bank's original
records had been destroyed.
In addition, historical records supply excel-
lent material for advertising and public rela-
tions, and can be used to cement the bond be-
tween management and stockholders. Interest-
ingly enough, the evidence reveals that in many
cases companies pay as much for an inadequate
"puff" story, in book or article form, as for a
sound professional history. Yet the "puff" story
carries no weight outside the company and serves
no purpose as an administrative reference in-
side the company.
Only about i % of a company's records can
properly be termed "permanent records." The
problem, therefore, is not how much is signifi-
cant but what is significant. What kind of rec-
ords should be preserved as of lasting histori-
cal and administrative worth? Particular care
should be taken to select material that highlights
why things were done, rather than just the usual
data on what happened or how it happened. In
fact, the biggest gap in business records gen-
erally is the answer to the question why.
A major task of records management is to train
industry to create and maintain "one file drawer"
of business history every year; that is about all
it takes to make room for films, recordings,
samples, drawings, and the like, along with writ-
ten and printed documents. Here are some of the
records that research has flagged as suitable for
this purpose:
Operating Department Records
Special reports on analyses.
Charts of plant layout, sample schedules, etc.
Photographs or drawings of the product.
Sample payrolls and statistical summaries.
Advertising and Sales Records
Summary sales statistics.
Samples of labels, packages, price lists, schedules, etc.
Samples of different types of advertising.
Special studies on advertising results.
Specialized Equipment
Mechanical devices and equipment are used
at all stages of the paper-work operation and are
essential to the development of professional
standards of records management.
Obviously it is important to know the equip-
ment to be purchased - how various makes and
models differ in price and performance. Wheth-
er company personnel have the opportunity to
check on installations in other organizations and
to go through all the steps required by a thorough
investigation, or whether they must depend on
outside assistance, the important thing is to as-
semble as complete and objective information as
possible.
Take the matter of microfilming, which is now
an old story as a way to save time and space and
to protect valuable documents. The figures in
ExHIBIT II show what a wide range of 'choice
in equipment there is for different needs and
suggest the form into which the information can
be put for ready comparison and analysis.
Or take the facsimile transmission of records.
Recently developed equipment makes use of the
photoelectric cell principle to transmit drawings,
charts, pictures, and typed or printed matter
across miles in a matter of minutes. This equip-
ment opens up new possibilities for decentral-
ized filing systems; they can now be located
"down the hall" or 15 miles away. Equipment
is put out by seven manufacturers in the United
States and England. As an example of the kind
of objective, analytical approach needed, here
is a checklist for comparing specifications and
performance, which NRMC worked out for its
own use :
Operating Features
Material for transmission Communication distance
General Records Document transmission Circuit privacy
Minute books and basic legal records such as charters and sensitivity Duplication processes
franchises. Maximum scanned area Operator training required
Executive correspondence on policy matters. Scanning rate Legibility of copy
Special or annual reports to directors or stockholders. Transmission time
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
Costs
Transmitter (purchase or Circuit 'charges
rental) Installation charges
Receiver (purchase or Recording paper charges
rental)
Administrative Decisions
While there are some problems of records
management which company executives can
delegate to specialists, there are other decisions
which they must make themselves. These deci-
sions. often come up in connection with the in-
stallation of a new records program, the question
of how long to keep the records, and the integra-
tion of the various systems and methods of rec-
ords management.
Installation
Scientific records management usually means
new methods, processes, and systems. Installing
them may be a simple matter involving changes
Records Management 59
example, the allocation of scarce resources. The
following case is illustrative:
Two years ago, as part of its records control pro-
gram, Oneita Knitting Mills cut down on its record
making by designing a new form - a single form
that performed all. these functions:
Acknowledged the order from the customer.
Requested information from the customer con-
cerning labels.
Served as invoice to the customer.
Served as shipping ticket.
Provided manufacturing specifications to the
plant.
Recognizing the need for consolidating forms
was simply a matter of common sense. But design-
ing a suitable form required specialized knowledge,
and deciding whether to use multilith, master sten-
cil, spirit master, or a multicarbon record required
judgment. Thus, in comparing the four alterna-
tive processes, relative weights had to be assigned
to cost, speed, quality of reproduction, and so on;
EXHIBIT II. USEFUL INFORMATION FOR SELECTION OF 16 MM:. MICROFILM CAMERAS
9" X 6"
Check
M" X sr"
87A" X Z4"
Kind of feed
Suggested
reduction
ratio
Average
images
per hour
Equipment
purchase
Automatic
24-40/1
2,500
Automatic
24-40/1
2,500
Automatic
24-40/1
2,000
Automatic
24-40/1
Automatic
24-40/I
By hand
17-24/1
By hand
17-24/1
cost
$45o-$3,300
$450-$3,300
$45o-$3,300
$450-$3,300
$450-$3,300
$450-$3,300
300
$450-$3
Personnel
cost per
image
$o.0009
$0.0008
$0.0010
$0.0010
$o.ooo6
$0.0o70
,
$0
00
0
Total cost
per image'
$0.0014
$0.0014
$0.ooz31
$0.0021
$o.oo15
$o.oi6o
.
7
$0.0160
Rental cost
of equip-
ment (per
month)t
$37-$76
$37-$76
$37-$76
$37-$76
$37-1176
$37-$76
$37-$76
" Cost of developer and paper.
t Alternative to purchase.
SOURCE: National Records Management Council, Technical Information Service.
only in the way things are done, or it may neces-
sitate new machinery and facilities. In either
case, executives in charge of the operation will
ordinarily need to choose between different al-
ternatives. There is no "package solution" to a
given records problem. The same kinds of con-
siderations usually bear on management's deci-
sion as in the case of other problems - for
and then these had to be balanced against the fact
that the company already had a multilith machine
with personnel trained to operate it. This was obvi-
ously'not a task that fit into a formula.
Retention of Records
Once the record-making processes are estab-
lished, the question arises : How long should the
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
60 Harvard Business Review
EXHIBIT III. RECORDS RETENTION SCHEDULES REQUIRED BY FOUR FEDERAL AGENCIES
(In years unless otherwise noted)
Federal
Interstate
Federal
Civil
Communications
Commerce
Power
Aeronautics
Commission
Commission
Commission
Board
Payroll registers
Vouchers
Plant
Permanent
Permanent
Permanent
Permanent
Other
10
15
IO
6
Accounts receivable
Accounts payable
6
Bank deposit books
Bank reconcilia-
tion + 3 months
3
3
Optional after
3
transfer to
other records
Materials and supplies
3
3
3
Engineering records
Completion
Completion
Completion
+ 10
+ 6
+ 6
Optional
Termination
Termination
+ 3 + 2
records be kept? Here again judgment should
be used. However, no intelligent decision can be
made on whether or not a specific record should
be kept - and for how long and where -
without having reliable and comprehensive in-
formation on (a) the legal requirements, (b)
comparable company data, and (c) the com-
pany's own experience.
Q Legal requirements - Assembling data on all
legal requirements to answer individual company
needs is a huge but straightforward job. The Na-
tional Records Management Council now has in its
files the complete index to every federal record-
keeping requirement. (It has not yet established
guides for both federal and state legislation on rec-
ords.) EXHIBIT III graphically indicates the vari-
ance in retention periods required by different fed-
eral agencies for the same type of record.
Q Comparable company data - In addition to
abiding by the law, companies, like individuals,
very often conduct themselves the way their neigh-
bors do; they follow the customs. So it is useful
to know what other companies are doing. (The
NRMC maintains a complete file of comparable
company data from controlled record-keeping in-
stallations.) However, such information should
be used as only one of the guides for assisting
management in deciding on records retention; and
it must be kept in mind that when compiled on a
broad scale, as by a business or professional associ-
ation, it may reflect what is being done rather than
what should be done. The fact that three authori-
tative banking groups differ widely on the records
retention schedules that they recommend, as shown
in EXHIBIT iv, indicates the necessity for taking
the pattern of any other organization, even in the
same line of business, with a grain of salt.
New York
Washington
National Association
Bankers
Bankers
of Bank Auditors
Association
Association
and Controllers
Personnel
Attendance records
4-10 years (for time
record card)
Permanently (unsuccessful
10-25 years after
leaving service
applications optional)
Commercial deposits
Deposit tickets
12-20 years
7 years
15 years
Trial balances
Until purpose is served
3 months-2 years
6 months
Loans and discounts
Loan applications
3-IO years
Until purpose is served
Optional
Note of discount tickler
2-12 years
1 o years
2 years
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
Records Management 61
Q Company's own experience - Legal require-
ments and comparable company data are not suffi-
cient in themselves. They must be weighed against
the company's own administrative experience.
There may be, for example, a six-year statute of
limitations on cases involving accounts receivable.
In actual practice, management may decide to
take a calculated risk, to discard the records in
question after two, three, or four years. Or it may
break down the records by size of order and dis-
pose of material relating to orders under $25 after
two years, orders under $5o after three years, and
so forth.
The above discussion applies mainly to rec-
ords with legal overtones. There are many other
records that have no connection with statutes
Integration
In taking an objective, realistic view of paper
work, records management studies and treats it
as a whole, as an integrated operation. Only
when all operations involving a piece of paper,
from its creation to its disposal, are coordinated,
does the paper work proceed from beginning to
end as smoothly and evenly as a product moving
along the assembly line. Indeed, the ideal is a
continuous flow from start to finish, without un-
necessary and costly delays and reworking.
Compartmentalized, disjointed paper work
breeds duplication and layering of operations -
which, in turn, breed needlessly high costs. A
plant engineer would be pained by the idea of a
I. Design and consolidate forms.
2. Develop "correspondex" form letters and
I. Inventory and appraise all records in
office and storage
2. Establish "timetable" (records sched-
I. Streamline paper work in opening and
closing cases.
2. Develop guides to improve maintenance
pattern paragraphs.
Develop guides on papers to be created
and filed.
ule) on how long and where to keep
each type of record.
Design new records-center operations.
and use of case folder.
Establish quality control of social work
operations.
RECORDS "BIRTH CONTROL"
I. Elimination of filing in current folders of over 140,000 pieces of
paper a year.
2. Elimination of typing 20,000 letters a year.
3. Elimination of dictating into 20,000 cylinders a year.
Release of $2,625 worth of space (2,100 square feet).
Release of $5,000 worth of filing equipment.
Some 25 tons of records scheduled for sale as wastepaper.
Indexing and prompt reference to all records.
RECORD-PROCESSING CONTROLS
I. Elimination of posting and filing Io,ooo registration cards a year.
2. Development, for the first time in social work, of a functional case
folder to point up the status and history of each case.
Highlighting ? actual and potential sources of error, and providing
measurable guides on actual quality performance by both clerical
and social workers.
or regulations. In these cases, studies of actual
usage - that is, the frequency with which ref-
erence is made to records and the purpose of
such reference - point the way to developing a
tight retention schedule both for office and for
storage.
In any event, no intelligent decision about
the disposition of a record can be made until
all three factors we have mentioned have been
taken into account. Applying a "canned" solu-
tion to the particular problems at a particular
company may give some results but will not
realize the full potential of scientific records
management.
nut being screwed into place by one worker on
the line, then loosened and screwed in again by
two more operators. Yet this sort of thing is
common in paper-work processing, because the
flow of paper from creation through processing
to disposal is not viewed in its entirety.
The experience of the Department of Welfare
in the City of Detroit illustrates the possibilities
for achieving "better records at lower costs" when
the paper-work operation is considered as an in-
divisible whole. Although this was a program
for a municipal government, it is typical of
what can be achieved by a scientific approach in
any organization. The steps of the program set
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
 Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9
62 Harvard Business Review
up as a result of the initial audit, with results government. It is demonstrating capacity to im-
achieved, are shown in EXHIBIT V. prove paper-work operations by adding to stand-
ards of quantity, cost, and quality; and it is fill-
Conclusion ing the need for better business histories. Every
alert administrator who has not already done so
Records management has come of age as a will want to investigate its potentiality for his
practical science designed to help industry and own organization.
LL records are made by human beings and read by human beings. Differences
Ain personalities naturally affect record-keeping systems. In order to achieve
maximum usefulness, a person designing a report must remember the strong points,
the weak points, and the prejudices of the man or men who will read it. Also, the
human weaknesses of those who make out the report should be kept in mind. . . .
The personalities involved . . . affect the form of the records more than they
affect the number of records. . . . It is altogether proper that these forms should
vary from company to company. The purpose of a record-keeping system is not to
please experts but to present facts in understandable form to the individuals who
are to use them. Since no two minds work exactly alike, a form which is best for
one person may not be best for another. As long as variations in the reports and
records stay within the bounds dictated by common sense, each company can
profit by concentrating on getting the facts across rather than adhering rigidly
to standard forms. . . .
Inevitably there is no sharp borderline between mental and written control but
rather a broad area in which either may be used. For the benefit of those who are
wondering whether a certain type of information should be in written form, a
few criteria are listed :
i. Inefficiency. Records should be used when it is obvious that waste motions
are resulting from mental or visual control. If men or machines, for instance, are
frequently idle, a written schedule may be the answer.
2. Inaccuracy. When it is noticed that the information obtained verbally or
mentally is not sufficiently accurate, it may mean that the stage has been reached
where records are necessary.
3. Lack of Time. When the executive finds that he has not had sufficient time
to look up certain information himself, it may be a signal that the information
should be sent to him in the form of a report.
4. Forgetfulness. When lapses of memory concerning orders or information
start to become frequent, written records may be required.
Under any circumstances, records should be kept to obtain certain definite in-
formation, not for the sake of keeping records. An alert management does not
endeavor to set up a cost accounting system; it tries to find out what the costs
are. For this reason, standard systems have a limited application for manufac-
turers, except in the bookkeeping work and, to a lesser extent, in inventory control.
Paul F. Lawler, Records for the Control of Growing Manufacturing Enterprises
Boston, Division of Research, Harvard Business School, 1947, PP. 36-44-
Approved For Release 2001/05/08 : CIA-RDP74-00005R000100020003-9