Queuing up performance: how to apply a virtual queuing strategy in your contact center.

Virtual queuing is a strategy for reducing ASA and trunk/toll costs. The methods are not as well known as CRM, workforce management or IVR self-service. A new concept a decade ago, virtual queuing has come into its own as a proven technique for improving contact center performance. Now is a good time to examine this technology, understand how it works and outline what managers can expect if they implement a virtual queuing strategy in their contact centers.

[ILLUSTRATION OMITTED]

Callback Systems

Voice-mail-based callback systems have been available for many years. The IVR asks callers to leave a message and telephone number and announces that their call will be returned in a reasonable time period. Usually this is a few hours, although it's hard to know exactly, which is why reconnection rates with customers who use this service are so low. Productivity is lost when agents take time to listen to these messages as well.

Other agent-reserved callback systems save a customer's place in line and call them back when it's their turn. A screen pop with the customer's account information is presented to an agent who is prompted to initiate the callback. This preview process, as well as the dialing process, both falsely accrue as talk time, thus lowering the productivity of the agent. A system with FIFO (first-in, first-out) capability that allows callers to receive the callback in the same amount of time that they would have waited on hold is preferred by customers because it sets the best expectation and addresses the customer's needs as quickly as possible. Once the customer enters this virtual queue, he or she is afforded the opportunity to be more productive, whether at home or at the office. It is called virtual queuing because the software inserts a virtual placeholder in the queue.

How Virtual Queuing Works

From a technical point of view, the virtual queuing solution has to do a lot of tricky things such as monitoring call flow, anticipating agent availability on a skill-by-skill basis and interfacing with other systems in order to collect CTI information. All the contact center manager needs to know, however, is how virtual queuing affects call flow.

The virtual queue sits quietly and waits for a spike in call volume. During peak call volume, the customer is educated of the expected wait time and is promised a return call in the same amount of time as if he or she had waited on hold. When this option is selected, the system uses a virtual placeholder to mark the customer's place in line in the ACD queue. Moments before it is the customer's turn to speak with an agent, the system predicatively places the return call, connects with the customer and transfers him or her to the head of the line as a the properly skilled agent is concluding the preceding call. This approach keeps agents highly productive, because there is no lag between calls and no agent intervention required to process the virtual queued call.

If the customer chooses to wait on hold, he or she may get frustrated, hang up, and try again. In contrast, a customer waiting for a virtual queue return call cannot abandon. Therefore, with a virtual queuing strategy, the abandon rate is dramatically reduced and, as a result, so are the total calls offered to the contact center. This, in turn, has a positive effect on workforce management tools because of a reduction in repeat calls and the capturing of true, unique caller demand.

One question that people ask is, how often is the customer there when the system calls back? If the proper expectation is set with an accurate estimated wait time announcement, then the return call reconnection rate has been shown to be well above 90 percent. Good estimated wait time calculations are very important to a virtual queuing strategy, and it's important to note that there is no "one formula fits all" for contact centers. You may have slow-moving queues with long talk times, you may have fast-moving queues, or you may have both. This is likely why your ACD's calculation of expected wait time is accurate only some of the time. A virtual queuing strategy offers algorithm flexibility based upon the dynamics of your contact center.

[FIGURE 1 OMITTED]

Benefits Of Virtual Queuing

Virtual queuing addresses customer satisfaction because waiting on hold is a major source of customer complaints. According to a Call Voice study, 64 percent of American consumers feel that the companies with which they frequently interact are not very respectful of their time.

Virtual queuing eliminates this source of frustration by giving customers choices and the freedom to do other things while waiting until it's their turn to speak with an agent. Often, customers will comment that the system makes them feel like the company appreciates the value of their time. Contact centers using virtual queuing have also reported a decrease in customer venting and call durations. They also see a 10- to 12-point boost to service levels during peak times and increases in agent efficiency.

Implementation

Unlike plain old callback, virtual queuing must be integrated with contact center technology to ensure a seamless call flow and the best customer experience. Much of the difficulty has been removed by the emergence of APIs and adapters for a wide variety of ACDs, enterprise routers and workforce management software. A smartly implemented virtual queuing strategy preserves the current investment in existing systems. It should maintain load balancing and skills-based routing, preserve the CRM data associated with voice calls, provide integrated statistical data that allow agents and managers to see both live and virtual calls in queue, and improve the accuracy of workforce management tools because of a decrease in abandons and increase in first-call resolution. In addition, some vendors provide technology that offers virtual queuing for Web-based contact as well, which may be integrated with voice calls in a single, unified queue. Finally, your contact center is a dynamic environment and the solution should enable the call center manager to adjust parameters on-the-fly, without needing to consult a technician.

[FIGURE 2 OMITTED]

Key Metrics For Contact Center Performance

A contact center manager considering a virtual queuing strategy should know how it will affect performance and what the key metrics are:

Toll time. Eliminating hold time saves toll costs. At two cents per minute, the cost savings for a mid-sized contact center is typically $5,000 to $10,000 per month.

Abandon reductions. Abandoned calls are typically reduced 25 to 50 percent. Every abandoned call is a sign of customer dissatisfaction. More than 40 percent of call centers have an abandon rate of 1 to 5 percent (supportindustry.com, 2003).

Talk time reduction. Contact centers using virtual queuing report a decrease in venting and call durations, typically in the range of 5 to 15 seconds. This increases the call handling efficiency of the agent labor pool.

Average speed of answer. A 2003 study by supportindustry.com indicates that 25 percent of call centers answer a call in 6 to 15 seconds, and nearly 20 percent take more than a minute to answer calls. ASA is sensitive to the longer wait times encountered during unpredicted spikes in call volume.

Service levels. Some call centers report greater than 50 percent improvement.

Workforce management. Because virtual queuing reduces abandoned calls and repeat calls, it optimizes contact center efficiency and improves the accuracy of workforce management forecasts.

Of course, virtual queuing is not appropriate for every contact center. If the strategy is to answer every call on the first ring and the contact center can afford to over-staff in order to achieve a 100 percent service level, then virtual queuing will not be helpful. This approach is sometimes practiced by catalog retail sales businesses. The main consideration is how much unpredictable peaks in call volume adversely affect operations and customer satisfaction.

Case Study: Utility Uses Virtual Queuing To Master The Peaks

Let's take a look at an actual application of virtual queuing and its impact on contact center performance, cost control, and customer satisfaction.

Atmos Energy is one of the largest public natural gas distributors in the U.S., serving 1.7 million residential and commercial customers in 12 states. The company operates call centers in Texas and Louisiana. Together, the two call centers' 180 agents handle more than two million calls per year related to billing, collections, new accounts, service and emergencies. Call volumes, call durations and customer complaints reach their peaks during the winter heating season.

During the 2000 to 2001 season, which was the coldest winter on record for Atmos Energy, the cold weather, combined with tight natural gas supplies, acted to push up energy prices. The impact on the customer support was a 15-fold rise in calls, forcing Atmos Energy to set up additional call centers and use field office staff to answer and return customer calls. The overflow calls were handled by a manual return-call system in which unskilled staff recorded the information and promised a callback. The regular agents were inundated with calls, and to prevent callers from receiving busy signals, Atmos Energy contracted with its long-distance service provider to queue the calls until an ACD trunk line was available. Toll costs increased from $13,000 for a typical January to $600,000.

After this experience, the company decided to implement virtual queuing as a strategy for managing peaks in call volume without adding additional staff. The system included both hardware and software, and it was integrated with the ACD and IVR at each contact center through the use of software adaptors.

After installing a virtual queuing system, the company found that approximately 60 percent of callers accepted the option to receive a callback, and more than 97 percent of the callbacks were completed successfully. Atmos Energy contact center agents reported that customers were much happier after the system was installed. Because there was less "venting" about long hold times, the average handle times for answered calls decreased by 10 percent, which alone generated $163,000 in savings.

During the next peak heating season, Atmos Energy was able to reduce staff by the equivalent of 17 agents. Moreover, the cost per call was 39 percent lower than during the previous season, primarily because the toll costs from calls waiting in queue decreased by 93 percent. Judged against these cost savings, the virtual queuing system paid for itself in four months.

Eric Camulli is Director of Technology for Virtual Hold Technology.

If you are interested in purchasing reprints of this article (in either print or PDF format), please visit Reprint Management Services online at www.reprintbuyer.com or contact a representative via e-mail at [e-mail] or by phone at 800-290-5460.

By Eric Camulli

Virlual Hold Technology

Table 1. Actual examples of the impact of virtual queuing.

Consumer goods company
Service level improvement = 42 percent
Talk time reduction = 17 seconds
Abandon reduction = 52 percent
Toll minutes saved = 570,000 (per month)
Payback period = 4 months

Insurance company
Service level improvement = 56 percent
Abandon reduction = 58 percent
Toll minutes saved = 1,125,000 (per month)
Payback period = 4 months

Telecommunications service provider
Service level improvement = 64 percent
Talk time reduction = 21 seconds
Abandon reduction = 48 percent
Toll minutes saved = 605,000 (per month)