MIS and Telecommuting:
By Jack M. Nilles
Friends or Foes?
Benefits of Telecommuting
A noted analyst of and consultant for telecommuting strategy
outlines the ways that it can pay off--if you avoid the lurking conflicts.
Let's say that you've just gotten all your company's PCs and workstations
interconnected without glitches, including the ones that need to talk to
the mainframe for mission-critical purposes. After all that work, now your
boss wants you to set up connections to employees' homes. Does this situation
sound familiar? MIS, meet teleworking and its right arm, telecommuting.
Teleworking is any form of substitution of information technology
(IT, such as telecommunications and computers), for work-related travel.
That includes computer, audio and video conferencing between work sites
and such activities as telemarketing and call center operations. Telecommuting,
a subset of teleworking, refers to periodic work outside of the central
office, one or more days per week either at home or in a telework center.
The primary object of substitution is the daily commute to a workplace,
on either a part-time or full-time basis. Whatever you call it, these distributed
work patterns are beginning to have a noticeable effect on MIS operations.
Telecommuting is not a fad that will go away. Rather, its growth has been
proceeding according to our long-range forecasts for several years. A recent
study for Telecommute America!, a nonprofit organization in Scottsdale,
AZ, that promotes telecommuting, found that nearly two-thirds of Fortune
100 executives view telecommuting as not only good for employees but
also advantageous to employers. And the word is getting around. There are
more than 10 million telecommuters in the United States today, and the number
should double by the end of the decade--to more than one of every 10 workers
in the country.
Properly designed and implemented (those magic words), telecommuting works
for a great variety of employees. Essentially, a potential telecommuter
is anyone whose job involves solo thinking or communicating that does not
require face-to-face interaction. That amounts to about half the U.S. work
force--and the fraction will increase as computers and telecommunications
become even more sophisticated. The reason employers like it is that it
produces net benefits to them, which is also why telecommuters and the communities
they work in like it. Everybody seems to win, except possibly the downtown
building owners and shopkeepers and the automobile industry.
The Impact on IT
If you're the one in your organization who saw this advantage and convinced
the CEO to try it, you've probably worked out the IT implications. If not,
there are important telecommuting issues to consider.
First, not all telecommuting is technology-intensive. Some telecommuters
do nicely with just a phone line, paper and writing devices. In some organizations
with which we have worked, these low-tech users constitute as much as 30
percent of the telecommuters. Thus, they are essentially not a problem to
IS managers. At the other end of the spectrum are the full-time workstation
users who need lots of access to CAD/CAM databases and/or full-motion video.
For them, modem connections won't cut it and an integrated services digital
network (ISDN) line may be marginally adequate; asynchronous transfer mode
(ATM) may be the only acceptable solution. These cases require planning
Most telecommuters fit somewhere between the extremes. They use PCs but
not all the time. They need access to their office PCs, but "sneakernet"
may suffice for file transfer. If they do need LAN or server access, that's
most likely to be bursty: short transmissions at irregular intervals. If
you routinely monitor traffic on your system now, it's fairly simple to
anticipate such usage patterns. In fact, telecommuting may smooth out the
peak loads because, for example, the telecommuters may check on their e-mail
earlier in the day than if they were to wait until they got to the office.
Whatever the needs of telecommuters are today, you can be sure that the
technology demand--and network access needs--will increase in the future.
But that's the news whether or not you have telecommuters. The primary difference
between telecommuters and the stay-at-officers is the data-link location
between their computers. All in all, the flip answer to what they will
need is this: whatever they need now plus an added telecommunication connection.
There are other system issues, such as the central question of access to
the company databases. Which telecommuters need absolutely current company
data for all they do? How many can make do with slightly older data, perhaps
on a CD-ROM? How many can wait until they are in the central office for
Another key question is what sort of network is required to provide the
interface to the telecommuters. There are two basic telecommuting options:
work at home or at a telework center. For home-based telecommuting, IS should
provide enough phone lines to satisfy reasonable peak demands and enough
bandwidth per phone line to satisfy user needs. For most telecommuters,
28.8kbps modems will do this year. For those using imaging systems, requiring
video conferencing or needing access to large graphics files, ISDN may be
in order. Yet as equipment arrives that complies with the H.324 and H.263
video conferencing standards, conventional telephone service may still suffice
for telecommuters who need almost face-to-face interaction.
As for computers for telecommuters, many organizations have adopted the
strategy of providing laptops instead of desktop PCs, together with docking
stations in the central office. This looks wiser as laptops gain in capability
and drop in cost. One of our insurance industry clients provided its telecommuters
with duplicates of their home office desktop equipment.
On the other hand, some organizations ask users to pay to telecommute. Because
of budget constraints, few departments in the City of Los Angeles could
afford computers for their telecommuters. Most required that prospective
telecommuters provide their own PCs if they needed them, although the city
has arrangements with local superstores to provide corporate discounts to
city employees. Consequently, Bruce Roberts, the city's telecommuting coordinator,
says that lack of city-provided equipment has not been a major deterrent
to prospective telecommuters.
One way to reduce costs without eliminating the traditional office environment
is a telework center, which can vary in size from a few to hundreds
of workstations. If the center is wholly owned and operated by the company,
it can be treated like a normal remote site with local networking, a wideband
data link to the central facility and video conferencing. If it is a multiclient
center, the center operator may or may not have a variety of suitable options.
In any case, the key requirement for a telework center is that it be located
close to the homes of the employees who work in it.
Security has to be considered a basic element of the telecommuting network
design. Security precautions start with restricting access to the company's
data. A variety of technology solutions are available, ranging from a packet
filtering router at the network interface to elaborate systems with token
cards and multiple firewalls. As companies begin to use the Internet for
telecommuting support as well as advertising and customer support, security
issues will grow. But most of these considerations will result from the
more general Internet activities. Telecommuting will just be another component
of the package of network services you need to provide. Some organizations
with telecommuters, such as the California Franchise Tax Board (which collects
state income taxes in California), will allow some telecommuters to work
only from telework centers, where physical security can be maintained more
readily than in employees' homes.
When the City of Los Angeles was developing its telecommuting program, one
of the concerns was control of access to the mainframe that contained personnel,
payroll and other sensitive data. According to Michael Galvin, the IS coordinator
on the city's telecommuting task force, the issue was resolved by issuing
smart cards to deserving telecommuters. Further, all the telecommunications
connections went to a firewall server that used a call-back system to make
the working connection. Yet, as many city departments have developed their
own LANs and provided direct access to their telecommuters, the demand for
mainframe access has not been as severe as was anticipated.
The most critical component of a good security system is not technological:
the telecommuters themselves. The admonitions you give to in-office employees
may need added emphasis here. Don't write passwords down where your teenage
hacker can find them; change passwords regularly; use a password-keyed screen
saver; and so on. Yet, lest the fears of rampaging telecommuters keep you
up at night, remember that telecommuters tend to be more loyal to the company
and less likely to give you trouble than some of the nontelecommuters.
By far the highest barrier to acceptance of telecommuting is between managers'
ears. Effective telecommuting management requires that telemanagers focus
first on clearly specifying the tasks to be performed, then concentrating
on the results produced by their employees, rather than the work processes
used to produce them. You can't manage by walking around if there's no one
around to walk to. Yet many managers feel dependent on eyeball contact with
their employees as a prerequisite for effective supervision. This attitude
is based on the assumption that the workers won't work well if the boss
isn't there to keep them in line.
This is a valid assumption for some employees. But for these cases, the
more fundamental question is, Why did you hire them in the first place?
The best employees are people who know what they're doing, can communicate
effectively with boss and coworkers, and are disciplined enough to complete
an assignment without much direct supervision. That is, they are ideal telecommuters.
My experience has been that most telecommuters become even better organized
and more focused on results because of their telecommuting. This, and the
fact that the rate of interruptions goes down dramatically, is the core
of the performance improvements seen in telecommuters.
On the other hand, not everyone is suited to home-based telecommuting, even
part-time. Some people need to be near fixed files or equipment (like the
mainframe). Others don't work well without daily social contact. In many
cases, these people will still work well in telework centers. IS managers
typically have many unheralded telecommuters, such as systems analysts who
respond to 2 a.m. crashes and programmers living in the boonies. The Travelers
Companies began their telecommuting program in the mid-1980s to have access
to some talented programmers who lived too far away for a daily commute
to the Hartford, CT, headquarters.
One of the next most common concerns about telecommuting is that teamwork
will suffer. The notion of teamwork, or group productivity, is gaining prominence
as companies deal with evermore complex business environments. To assess
the impact of telecommuting on teamwork, it is important to step back and
review the process.
Most teams go through a regular series of interaction cycles as they pursue
their goals. First is the team startup period. Here, it is usually necessary
for the team members to get together physically as they define the team
goals, allocate responsibilities, and assign individual tasks and objectives.
The need for face-to-face interaction comes from the fuzziness of definition
of these tasks at first. It's possible that some form of electronic conferencing
can work even in these situations, but the usual approach is to suspend
Once the assignments have been made, however, the key productivity enhancer
usually is solo work, which can be done better by telecommuting. Further
team meetings may be necessary to review progress at milestones or to revise
objectives, but often even these meetings can be done electronically. The
point is that both meetings and telecommuting should be used when they are
the best productivity producers. One team of programmers for the City of
Los Angeles won a performance award by producing in three weeks of telecommuting
a program that they hadn't been able to organize for more than a year. The
reason? "Telecommuting allowed us to get the uninterrupted time to
think about the problem," says one team member. "We communicated
with each other by phone at least as well as we would have in the office."
Training and Helping
To make this new model work, telecommuters and telemanagers need two types
of training. First is the technology training that aids them in communicating
with each other and with the rest of the company and customers. If you are
introducing new technology, such as ISDN services, along with telecommuting,
this training is especially important. The telecommuters should be made
aware that they are more likely to be on their own while telecommuting and
consequently more alert during the training sessions.
Equally important, and even more so for supervisors, is management training.
Both telecommuters and telemanagers need to know the basics of working together
apart. Topics to be covered include establishing goals, recognizing results,
setting schedules, communicating effectively, getting organized and managing
time. Our years of testing telecommuters and telemanagers make it clear
that this training can be critical to success.
An effective help desk is more important to telecommuters than to people
in the office. Repeated surveys of telecommuters have shown us that there
are two prime sources of help in most organizations: the help desk and the
local guru. Naturally, the local guru is no longer local in a telecommuting
situation, so the help desk is often the only source of support. Yet telecommuting
can often hasten the learning process, to everyone's benefit.
For example, one local guru was constantly being asked by coworkers to help
them solve one computer problem or another. Usually they forgot the solution
within minutes after the guru left their desks. Since the guru was getting
paid for other things, this had a negative effect on his productivity, and
the forgetful coworkers weren't improving either. The crunch came when the
guru became a telecommuter; he no longer was available to repeat their fixes.
Within a few weeks the coworkers learned what the guru had taught them.
Both guru and coworker productivity improved.
Payoffs and Costs
Beyond the qualities mentioned above, there are bottom-line motivations
for an adopting company. Our research indicates that the productivity of
successful, one-or-two-day-per-week telecommuters averages 5 percent to
20 percent higher than their nontelecommuting colleagues--as measured by
their supervisors. Telecommuters use less-to-no central office space or
less expensive space, depending on how often they work at home and/or at
a telework center. Ditto for parking space. Many organizations, such as
IBM, Digital Equipment Corp. and the State of California, have adopted telecommuting
on the basis of space savings alone.
Telecommuters tend to take less sick leave than nontelecommuters, typically
two or more days per year. They are more loyal to the company; turnover
rates are reduced, often dramatically. (In my first telecommuting project--in
1973--the turnover rate went from 30 percent to zero. More generally, 20
percent or greater reductions are common.)
The productivity of their stay-in-the-office colleagues also goes up, due
to generally improved organization. Finally, many companies are in urban
areas where air quality regulations mandate reduced car use. Home-based
telecommuters usually do not use their cars for commuting or anything else
while they're working. Traffic congestion is reduced, energy is saved and
air quality improves, so the community benefits. We have found that, added
together, these produce a typical net annual benefit of from $6,000 to $12,000
per telecommuter for midlevel employees.
There are two primary cost areas for telecommuting. The first is startup
costs: planning, selection, training and technology implementation.
The second is operating costs: mostly telecommunications charges
and equipment maintenance. Most of the long-term costs of telecommuting
are in the domain of IS, if you include telecommunications in that category.
Additional computers, docking stations, telecommunications hardware and
software (and their installation) tend to be the dominant up-front costs
for large telecommuting programs. The primary additional costs are for training
and implementation project management.
The computer question is an important one. Foremost is the decision as to
who's buying. Most government agencies in this era of tight budgets restrict
telecommuting to those who can provide their own PCs at home. Most private
companies provide the computers for their employees. Usually it is not necessary
to duplicate equipment. Full-time telecommuters simply move their computers
to the telecommuting location. The majority--those who split time between
a telecommuting location and the office--may use the option of laptop and
docking station, so the duplication tends to be in monitors and docking
stations rather than entire systems. Where LAN-like performance is a requirement,
add the installation costs of ISDN lines.
Telecommuters get some financial benefits as well. These include reduced
operating costs for their cars, fewer trips to the dry cleaners and less
expensive lunches. But the most important benefit, repeatedly told to me
by telecommuters, is stress reduction. That comes from eliminated commutes,
of course, but also from the huge jump in work continuity (interruption
hiatus). Telecommuters are able to fit their work into the rest of their
lives, instead of vice versa.
Making It Happen
Successful telecommuting programs tend to have three main phases: planning,
demonstration implementation and rollout. Each has important implications
for IS managers. Rule One is to have upper management support, preferably
among several upper managers, before you begin. These managers need not
be telecommuting enthusiasts, but they should be willing to give objective
reasons why they will back expansion if the demo is successful.
Because telecommuting does not necessarily come naturally to all employees
and may require system design changes if computers are to be networked,
plans for both the demonstration and rollout phases are imperative. A primary
function of the planning phase is to establish success criteria, performance
measures and preliminary policies for telecommuting. That is, did we do
well, and how do we know?
For the demonstration phase, the key objectives for IS are establishing
the networking functional and technology performance requirements, scheduling
and costs. If your company is thinking of introducing telecommuting on a
broad scale, it is important to include a representative sample of organization
units in the demonstration. Some units may need extensive technological
support, others little.
Make sure those niggling details are considered, such as transparent (well,
glitch-free) access to the department network by laptop-equipped telecommuters.
If you're using a groupware product, make sure that it will support these
remote users. Try not to implement new forms or scripts at the same time
the demonstration participants begin telecommuting; you want to be able
to separate telecommuting from technology adaptation problems. A key cost
factor is the contention ratio--in this case, the number of telecommuters
divided by the number of telecommunications ports required to serve them.
The belt-and-suspenders approach would have one port per telecommuter, but
ratios much higher--and less costly--than that might suffice. They may also
help to constrain the security concerns.
Once the plan has been made as bulletproof as possible, the technology is
in place and all the participants have been trained in both technology use
and management practices, you're ready to go. The demonstration phase should
be used to fine-tune the technology and training needs and to provide critical
data for the rollout. Don't be hasty. Telecommuting requires some adaptation
time. I usually recommend at least a year for the demo; nine months is the
minimum. Too many things are still in flux in the first several months after
implementation for reliable follow-on decisions to be made.
The demo phase is also the time for assembling the core of the rollout plan.
In essence, repeat the processes of the planning phase, but refine the numbers
with the data derived during the demo. Central to this task is the help
desk. Make sure that the help personnel realize that this is an important
project, give it full support, keep careful records of the problems that
come up and report them immediately to the implementation team.
Telecommuting works and works well when properly implemented. Conversely,
poor planning and failure to get management's backing can be dangerous to
your career, if not your health. If your actions can result in a net annual
benefit to the company of, say, $8,000 per telecommuter, you may deserve
a hero button, if not a percentage. Even better, you may actually get recognition
for being a stellar innovator. Besides, if you become a telecommuter yourself,
you may be forced to figure out what to do with all that new-found discretionary
time you have.
Jack M. Nilles, who coined the terms teleworking
and telecommuting in 1973, is president of JALA International, Inc.,
and author of Making Telecommuting Happen: A Guide for Telemanagers
and Telecommuters (Van Nostrand Reinhold, 1994). He can
be reached at firstname.lastname@example.org.