In response, the department installed a voice-mail system. It soon learned, however, that the system only alleviated a portion of the problem. Although voice mail ensured that callers didn't remain on hold for lengthy periods of time or call back busy numbers continually, it didn't make it any easier for them to reach the right person, nor did it guarantee speedy call-backs. "A machine simply didn't present the kind of service we wanted to provide," says Kathleen Needham, manager of HR information systems, who helped procure the voice-mail system. "We were dealing with everyone in the hospital, including department heads, and they deserved a more personalized service."
Besides lacking the personal touch, voice mail was inefficient because it requires at least one call-back and in some cases an unwelcome game of voice-mail tag. "The problem grew on itself," says Jim Sponsler, acting vice president for HR. "Recruiters had to spend time on the phone tracking down people who left messages. Because they were spending so much time doing that, they couldn't receive their incoming calls."
It was clear to everyone in the department that a problem still existed. However, a satisfactory solution eluded them. Sponsler knew that one way to fix the problem would be to increase staffing until the situation improved. But today, few hospitals, or any corporation for that matter, can afford to use the shotgun approach. To make the most efficient use of resources, Sponsler decided he needed to quantify incoming and outgoing calls by time of day, area and category. "Then we'd know exactly where the bottlenecks were, and we could determine the most effective areas in which to add or reassign staff."
Recruitment and benefits inquiries create logjams.
To help him study the HR department's phone troubles, Sponsler brought in Joe Koczan, a management consultant who specializes in the health-care industry. Under contract with the hospital, Koczan spends much of his time working on efficiency projects. Between spring and fall 1992, Koczan interviewed the HR personnel. Through these interviews, he was able to determine that the phone tie-ups were most severe in two areas: recruitment and benefits.
The tie-up in recruitment was caused by a constant abundance of calls. Because the hospital hires or transfers 43 people monthly on average, it receives approximately 800 applications a month. People frequently follow up with phone calls to verify application receipt, confirm interviews and check position status.
In the benefits area, the personnel could handle the number of phone calls but not the number of technical questions. Callers usually got through to a benefits person on their first calls, but often had to wait for a call back from a technical expert to get their questions answered.
HR staffers told Koczan such things as, "There are too many calls," and, "There aren't enough people to answer the phones." To find exact solutions to the problem, Koczan needed to quantify these statements. He decided to conduct a formal study of the length and type of calls coming into the HR department.
Traditional time studies require consultants with stopwatches and notepads to follow employees around and note the start and stop times of each activity. That procedure tends to be intrusive and very expensive. Koczan chose instead to use a bar-code system that would enable workers to do their own tracking, eliminating the high fees of the traditional system.
Bar-coding first came to use in this country in the wholesale and retail industries to track products. In the past decade, the technology has expanded into other industries, including the medical industry. Some hospitals, for example, put bracelets containing bar codes on patients' wrists. Instead of nurses noting down medications on the patient's chart and transcribing the notes at the end of the shift, they simply wand the codes on the medicine bottle and on the patient's bracelet.
The adaptation of bar coding for time studies and training-need assessments was the brainchild of Bill Lykes, president of Arlington, Virginia-based Work Management Institute Ltd. (WMI). With Lykes' bar-coding equipment, employees wand all of their activities during the day. Instead of having one consultant per staffer remain on premises for an entire period of a study, only one consultant visits the site for just two days—one day to set up the system and train the users and the second to debug the system. Employees send study results electronically to WMI, which generates a report. "The HR department seemed a perfect fit for bar-code technology," says Koczan. It could provide an easy and inexpensive method to categorize all incoming and outgoing calls.
In October 1992, Koczan brought WMI on board to implement a bar-coded study. The first thing they did was establish the categories of calls that would be bar coded. It took several iterations to get good sets of categories because there's a natural tendency to be too specific. For example, Needham says that the first list in the benefits area filled four or five pages and included such categories as "calls about paid time-off," and "calls about the value of shift differential." To make it easier for the HR personnel to find appropriate bar codes quickly, WMI suggested they limit their categories, using phrases such as "calls about wage policies" to encompass several of the first, narrower category descriptions.
After the lists were pruned, Koczan sent them back to the employees for one final check. Each worker had to determine whether the list covered all the types of calls that the area received and if the categories were specific enough so that no calls could fall into more than one category. WMI then created sheets of paper that contained labeled bar codes for each type of call in each area.
A WMI staffer came on site to train the users. Training lasted one hour, and then for the rest of the first day, users experimented with the system. The next day, HR employees had a dry run, in which they used the system as if they really were collecting data. The goals were to get users accustomed to wanding while also verifying that the call categories were comprehensive and didn't overlap.
The time study largely confirms suspicions.
Each HR worker toted a credit-card-sized Videx TimeWand I bar-code reader and his or her area's sheet of bar codes for two weeks. When a worker received or placed a call, he or she wanded the bar code that represented the topic of the call once when each call began, and again when the call was completed.
Needham's initial reaction was that it would take too long to determine the type of each call. "We often had to be into the call for a few minutes before we could code it," she says. However, after practice, she was able to predict call categories quickly; wanding became second nature.
The list of bar codes included a couple of error codes. One signified that the previous scan was an error and allowed the worker to scan the correct code. The other error bar code signified that the worker had forgotten to scan a new activity. When this occurred, the worker then made a correction manually on a data sheet.
At the end of each day, the TimeWands were inserted in a Videx Multiple Recharger, which recharged the wand and sent the data via modem to WMI's desktop computer for analysis. The WMI software, called Quixolve, checked for obvious errors such as activities that took much longer or shorter than the average.
Two weeks after the survey was completed, Koczan received a report from WMI on disk and paper. It included such information as number and average duration of calls by category and area. Koczan asked workers if they believed the reports were accurate. All agreed that they were.
Analysis of the study confirmed much of what Koczan had suspected. For example, it revealed that the benefits area received more technical calls than could be handled by the single technically skilled employee in the area. Fortunately, the study also showed areas in which technically skilled personnel were underused. Sponsler solved the problem by transferring a person with technical knowledge to the benefits area without sacrificing service in the area from which she came.
The report also revealed that the majority of calls received by recruiters were for appointments. Having highly trained recruiters take up their time with scheduling calls was a waste of resources, says Sponsler, so he hired an administrative assistant to handle that task.
In addition to confirming what initial interviews suggested, the report revealed other problem areas. It indicated that the department receives a preponderance of questions regarding newly instituted policies. To reduce the number of these calls, the department now includes informational pamphlets with pay checks. As a result, according to Needham, the number of those types of inquiries have dropped.
The study also found that many calls to information systems were from outside sources asking for employment-verification information. The response is always the same: The department only can provide information upon written request from the employee. Although it isn't possible to entirely eliminate those calls, the department determined that a message in the area's voice mail explaining the policy can significantly cut them down.
More changes to increase the efficiency of the department are still to come as a result of the time study. The project, not including the implementation of new systems or new positions, cost the department $25,000, $5,000 of which went to WMI and the rest of which covered Koczan's fees. Sponsler doesn't cost-justify it, partly because he can't predict just how much more efficient his department will be once all the changes are implemented. "The problem had to be fixed," he says, and couldn't be quantified in terms of dollars. Moreover, he says, the advantage of the bar-coded call survey is that, "Now when we plan changes, we're working with facts, not guesses. We now know exactly how many calls each employee receives and what kind of calls each receives. All that is tremendous help in planning."
Larry Stevens is a free-lance writer based in Massachusetts.
Personnel Journal, April 1994, Vol.73, No. 4, pp.109-115.