July 2000

BY JIM HOGAN, PIPKINS, INC.

As call centers become more technically complex and agents become more specialized, staffing becomes an increasingly critical issue. Having too few agents available drives away customers (a dangerous practice with the competition only a click away), while having too many agents is costly for the company. Additionally, today's tight employment market makes retaining skilled agents a top priority. These trends have made selecting the right workforce management software program more important than ever -- in fact, it's mission-critical.

Fortunately, workforce management software has responded to the market and is evolving to meet these needs. More sophisticated technologies are being developed to optimize call center staffing, increase employee satisfaction and improve the bottom line. These enhanced technologies are most apparent in two critical areas: call volume forecasting and agent scheduling.

Forecasting call volumes and staffing requirements requires sophisticated algorithms, such as the Erlang-C equations employed by most workforce management software programs. Erlang-C (developed in the early 1900s for telcos) calculates the number of agents needed to staff a call center based on the total number of calls received historically and the skill sets of the agents. However, this algorithm assumes that the call center can physically answer every call that comes in (no caller will receive a busy signal), and that customers will wait indefinitely for their calls to be answered. In reality, customers will eventually abandon (hang up) if kept waiting too long. Other callers will receive a busy signal when the available queues are filled. The end result is that the center is always overstaffed. A more sophisticated formula is a modernized version of the Erlang-C forecasting algorithm, which takes into account both busy signals and call abandonment. This results in a more precise calculation of the number of agents required to handle the projected call volume, eliminating the overstaffing inherent in the traditional formula.

Pattern recognition, another key forecasting advancement, allows users to identify abnormalities in the underlying historical data that can be attributed to a certain event. Each time that event occurs, the forecasted call volume will automatically be adjusted to take into account the historical differential associated with that event. For example, a call center may typically handle 4,500 calls on a Monday morning in July, but if that day happens to be Independence Day, the volume may drop by 40 percent. Pattern recognition technology automatically ensures that the right number of agents will be scheduled on the holiday to handle the reduced call volume, allowing other employees to enjoy a picnic rather than sit in the call center waiting for the phone to ring!

More sophisticated logging and monitoring systems have also improved the granularity with which volume can be predicted. Where service levels were once defined for a full day, they can now be defined in 30-minute or even 15-minute blocks. This enables more efficient scheduling of breaks, meetings and training sessions, further optimizing agents' schedules.

The latest advancements in scheduling capabilities can do more than improve the company's bottom line -- they can also enhance agents' job satisfaction, increasing performance and reducing turnover. Multiskilled agents can be scheduled for multiple activities during a shift, with the software ensuring that all activities are covered at all times. Additionally, scheduling options are more flexible than ever. Agents can stipulate preferred start and end times or preferred days off, and employees can be rotated through less desirable shifts. All of these factors result in increased job satisfaction for agents and reduced turnover, therefore decreasing training and hiring budgets.

The more innovative workforce management providers are not simply adding bells and whistles to their existing systems. They are responding to related advancements in technology by incorporating Internet and wireless technologies. For example, agents can use the Internet to request changes to their schedules and even swap schedules or shifts with other agents, making the workforce management system an interactive environment between the supervisors and the agents. In a wireless environment, the supervisor is unshackled from his or her desk, equipped with a hand-held device and free to roam the call center floor, while browsing into the system to access real-time information and respond to any changes in trends.

Another advancement is the ability to forecast and schedule agents for work other than the traditional inbound call, such as e-mail or Web chat. Many companies are seeing a significant increase in customer communication via e-mail, the handling of which requires staff members to have specialty training and skills. As e-mail volume increases, the need to schedule these agents efficiently will increase as well.

Exactly how much impact can this new technology have on a call center's bottom line? One prominent Fortune 500 company estimates that they were scheduling 2,500 agents per week at 80 percent efficiency when they decided to upgrade their workforce management solution. Once the software had been deployed enterprisewide, scheduling efficiency improved to nearly 94 percent. Each percentage point increase represents an increase in the bottom line of approximately $650,000, making their efficiency gains worth somewhere in the neighborhood of $7 to 8 million annually. Granted, results for smaller companies will be less dramatic, but a percentage point increase in efficiency in a small to medium-sized call center can still represent tens of thousands of dollars. In most instances, a workforce management package has a payback of less than one year based on staffing efficiency alone. Add to that the increase in customer satisfaction and the increase in employee job satisfaction, and it's obvious that it pays to keep up with the latest in workforce management software.

Jim Hogan is senior consultant and manager of customer care for Pipkins, Inc. He has been project manager for over 20 installations of Maxima Advantage, Pipkins' workforce management system.

[ return to the July 2000 table of contents ]