Out With The New, In With The Old: A Look
At Scheduling Alternatives
BY BRIAN SPRAETZ, IEX CORP.
A long time ago, Darwin proposed that nature had an innate process for
determining the best choice between options. Natural selection, in his
view, worked to ensure that species best equipped to survive would
flourish, while those less capable of dealing with the harsh realities of
life would disappear. Life moved onward and upward in lockstep with the
Theory of Evolution, innovating and improving in order to survive in an
increasingly competitive environment.
There are, however, some notable exceptions to this generalization.
Sharks, alligators and sea turtles, among other species, remain seemingly
immune to change. Perhaps change is not an absolute requirement for
survival. Maybe simply maintaining the status quo is sufficient to get by'
but for how long? In the area of contact center scheduling, a strikingly
similar parallel exists. Just like sharks, many organizations cling to an
agent scheduling method that is quagmired in evolutionary muck'a method
in which agents bid for work assignments through a process called schedule
bidding. Is this status quo approach sufficient to ensure survival in
today's business environment? Before rushing to a judgment based on the
ever-popular 'newer is better' theory, it's worth taking a closer
look at the pros and cons of schedule bidding and a popular alternative
approach, preference scheduling.
The Evolution Of Scheduling Systems
Scheduling of employees was not a major issue until the turn of the 20th
century, at which time the nation's telephone system began to grow
rapidly. Before automated switching equipment was invented, telephone
callers had to be connected manually by operators. Telephone companies
began to experience problems with staffing and scheduling as hundreds,
even thousands, of operators were needed to handle the increasing call
volumes. A variety of approaches were conceived to address this scheduling
problem, most of which relied on hands-on experience, seat-of-the-pants
judgments and more than a touch of luck. Eventually, Mr. A. K. Erlang and
his magic formulas appeared in the early 1920s.
Still in use today, Erlang's equations for determining the number of
agents required to meet a desired service level, with a forecasted calling
volume, provided mathematical rigor to what had largely been a guesswork
approach to the scheduling problem. With a scientific method for
calculating the number of workers needed during each specific time
interval of the day, the focus switched from trying to figure out how many
people were needed to finding an efficient way to schedule and allocate
working assignments. Enter schedule bidding.
The process where employees select, or bid, on the working shifts they
prefer from a master list of all possible schedules was seen as a fair and
equitable method of allocating assignments, at least to the employees with
the most seniority. Schedule bidding is basically a two-step process in
which a set of schedules is created to cover the forecasted workload
requirements and then those schedules are presented to employees for
bidding in some sort of pecking order such as seniority.
Giving employees the ability to choose among the alternatives provided
workers with a feeling of control over their assignments. Even when given
only a 'lesser-of-evils' decision, the fact that they were personally
making the decision, rather than having it dictated to them, helped foster
an environment conducive to improved job satisfaction. Unfortunately,
Erlang's equations and the schedule bidding process did nothing to
address the issue of what schedules were best for the organization in
terms of meeting service goals and optimizing resources. Enter modern
workforce management systems.
With the advent of mainframe computer technology in the 1960s, larger
organizations began to attack the scheduling optimization problem. The
development of in-house solutions, coupled with the PC computing
revolution of the 1980s, eventually led to the introduction of packaged
workforce management solutions in the early 1990s, which continued to
evolve into the robust applications available today. These marvels of
applied technology, capable of considering thousands of possible
scheduling combinations in seconds, optimize the matching of scheduled
resources to forecasted volumes, lowering staffing costs without
compromising service levels.
What appeared to be the sole remaining advantage of schedule bidding over
the emerging solutions, allowing employees to exert a degree of control
over their working assignments, was overcome through the addition of agent
preferences to the automated scheduling process. By predefining an agent's
personal preferences about schedule tradeoffs, total automation of the
schedule generation and allocation process was made possible in a way that
addressed the individual desires of employees while still conforming to
the operating goals of the enterprise. Schedule bidding as an allocation
method was surely destined to follow in the footsteps of the dinosaurs.
Sharks predate dinosaurs by thousands of years, and more important,
outlasted them by thousands more. Like the shark, the decades-old method
of scheduling bidding still flourishes in today's contact center.
Although especially prevalent in a few vertical industries, such as
airlines and car rental companies, its popularity spans a wide spectrum of
service industries. There must be some underlying rationale for this
First, there are some downsides to schedule bidding as a work
allocation process. In a bidding environment, schedules are typically
fixed for much longer periods of time -- months or even years in some
cases, as opposed to the agent preferences approach were they are often
prepared only a week or two in advance. These longer time frames do have a
practical basis for their existence. Schedule bidding's two-stage
process, often paper intensive and requiring a high level of manual
involvement, is not something an organization would want to repeat often.
Longer time horizons, however, increase the possibility that staffing
changes caused by employee turnover, or changes in the volume of contacts
coming into a center, result in an increasing number of coverage problems
as time progresses. The heavy reliance on manual involvement in the
bidding process also leads to higher labor costs and a greater potential
for human error than a fully automated approach.
Resistance To Change
Given the seeming superiority of using automated preference-based
scheduling in a contact center, why does schedule bidding continue to
remain in use? One possible explanation is found in the theory of
organizational inertia, or the resistance of an organization to change.
Even in the face of convincing and viable reasons to change, enterprises
often elect not to do so. This resistance to change can result from a
highly ingrained corporate culture or by being constrained in the ability
to change due to labor management issues. Whatever the underlying reasons,
many organizations find change a difficult concept to embrace.
Employees' active involvement in choosing work assignments in the
schedule bidding process presents another possible reason for resisting
change. Active involvement fosters a greater perception of control.
Although arguably a false perception compared to the use of agent
preferences, it is still a real concern when it comes to the impact on
employee morale. Letting employees bid on schedules, taking their fate
into their own hands, is often a more attractive alternative than spending
supervisory time compiling and presenting proof that a computer really did
provide the most desirable schedule for a particular agent.
A New Era For Schedule Bidding
Today's version of the shark does differ in some ways from its
pre-dinosaur-era ancestors. Change has occurred, even if it is not readily
apparent. Likewise, the computational power and sophistication of modern
workforce management solutions has brought about underlying improvements
in the schedule bidding process. These improvements provide organizations
using schedule bidding opportunities to achieve additional benefits
without switching their scheduling process.
Improved forecasting accuracy, combined with the ability to quickly
evaluate multiple scheduling alternatives, generates schedule sets for
bidding that better optimize resource use and meet service goals. Closer
matching of schedules to forecasted volumes reduces the amount of time
supervisors must spend manually manipulating and adjusting schedules. This
time can be better spent evaluating the cost impact to the organization of
alternative scheduling policies through 'what-if' analysis, or looking
for other areas of potential improvement.
Modern schedule bidding systems replace the use of paper-based forms
and reports with electronic versions, speeding up the bidding process,
reducing the potential for human error and making it all more convenient
for supervisors as well as agents. Once bidding is closed, agents'
selections can be evaluated against bidding rules and schedules
automatically assigned without the need for supervisory involvement.
Schedules that have not received any bids can be easily identified,
enabling prompt action to resolve potential coverage issues. Agents who
have not input bids within the allowed time window can be similarly
identified, allowing proactive investigation into the reasons.
The effects of agent turnover and changes in contact volumes occurring
over those longer scheduling time frames typically associated with
schedule bidding can be alleviated by alerting supervisors to potential
problems ahead of time. This early warning provides time to seek
alternatives and resolve these issues before they become critical.
Mixing The Two Approaches
The debate over which scheduling approach is best has been going on since
the advent of the preference-based models. This debate is likely to
continue for a long time, quite possibly never producing a clear winner.
Perhaps the greater benefit resides in a combination of the two
approaches, where the strengths of each are leveraged.
For operations using schedule bidding, the effects of agent turnover
and changes in contact volumes between bid cycles could be addressed quite
effectively by using preference scheduling to fill in the gaps with newly
hired agents, part-time agents or contract labor. Schedules that did not
receive bids could also be automatically assigned by preference
scheduling, using either the same labor pool or agents who were eligible
for bidding but did not place bids.
Conversely, in a preference-based scheduling environment, bidding could
provide an effective means for fair and impartial allocation of overtime
work assignments. For organizations supplementing in-house staff with
contract labor, long-term, fixed-schedule contracts could be allocated
through the bidding process while using preference scheduling to maximize
the use of the in-house agents. In a similar manner, long-term fixed
schedules could be offered to a center's high performing agents as a
reward component of a strategic employee retention program.
Contact center operations spanning multiple sites may be forced to deal
with differences in work provisions or labor management issues between the
sites. The requirements for some sites may be best addressed through the
use of schedule bidding while others may be more suited to a preference
scheduling approach. In cases such as this, the ability to combine
approaches within a single workforce management system enables centralized
planning, reporting and analysis across the enterprise as opposed to
managing these sites individually.
Like the shark, schedule bidding has survived the tests of time with
relatively little change. Fine tuning the bidding process continues
through additional automation and improved analysis and reporting
features. Workforce management systems supporting both types of scheduling
approaches no longer force contact center organizations into making a
decision of using one method exclusively over the other. Further, a hybrid
approach combining the best of both processes, and offering potential
benefits greater than either approach used alone, may be the ultimate
Brian Spraetz is the director of marketing for IEX Corporation. He
has over 15 years of experience in bringing high-technology solutions to
the customer service and contact center markets. IEX is a provider of
workforce management and intelligent call routing solutions.
To The November 2001 Table Of Contents ]