Inside this editioni

Transcription

1 WPP Summer 2005 A quarterly publication of the Society of Workforce Planning Professionals Workforce Planning Inside this editioni N S I D E WFM Survey Results...2 WFM Survey...3 Three Steps for Attacking...4 Adherence Problems It s Budget Time. Do You Know Where Your Shrinkage Is?...6 Excel Tips for Workforce...8 Planning Professionals Our Sponsors...8 Ask the Workforce Wizard...9 Events Calendar...10 Managing by the Numbers...11 WFM Success Story...12 Does Accuracy Matter?...13 Industry News...15 Join SWPP...16 Society of Workforce Planning Professionals 6508 Grayson Court Nashville, Tennessee SWPP Announces Workforce Management Certification Program The Society of Workforce Planning Professionals (SWPP) announces the creation of a certification program for workforce management professionals. The certification program, which will be available in late Fall 2005, is open to anyone in the industry. The certification allows the recipient to use the following title after his/her name: Certified Workforce Planning Professional (CWPP). Since SWPP was created, one of our goals has been to create a certification for workforce management professionals, said Vicki Herrell, Executive Director of SWPP. We are so excited about this program and what it will mean to our members to have a professional certification. The hardworking people in this industry deserve to be recognized for the professionals they are. The workforce management certification is composed of four parts, which includes three tests and a project. Satisfactory completion of all four elements results in certification. The tests cover three areas of expertise: Planning and Strategy Staffing and Scheduling Managing Daily Staffing and Service Certified CWPP Tests are taken on the web and must be proctored. Each test can be taken separately, or, if desired, all three tests may be taken at one time. Students receive test results upon completion of each test. The project is designed to be a real-life example of workforce management skills to demonstrate the application of knowledge. Sample projects will be created and approved by the SWPP Board of Advisors to make it easier for students. However, students may also submit proposals for other projects as well. The cost for the certification is as follows: $350 for Non-Members $250 for Members Study materials for the certification will be available on the SWPP website. However, there is no requirement for any student to attend any training. Applicants with experience in the field may have gained their knowledge through a variety of sources and it is only important that the knowledge be demonstrated through the certification process. SWPP s education partner, The Call Center School, is responsible for the development of the competencies and tests. For more information about this exciting new program, please visit our website at Professional

2 W F M S u r v e y Workforce Management Survey Results This article details the results of the most recent SWPP quarterly survey on critical workforce planning topics. In this survey, which focused on workforce optimization (WFO), approximately 200 call center professionals representing a wide variety of industries participated and provided insight into numerous workforce planning questions. Participant Profile The largest percentage (46%) of the participants was from large call center operations with over 500 seats, followed by 26% with agents. Twenty-eight percent had less than 100 agents. All types of call center operations were represented in the study, with the biggest percentage representing financial, insurance, and outsourcing. Current Tools The largest percentage (90%) of the respondents currently has a workforce management system in place. Eighty-three percent have quality monitoring systems, while 44% have hiring and recruiting tools, 37% have dashboard reporting and an e- learning system, 30% have agent performance analytics, and 19% have an automated customer satisfaction measurement tool. The smallest group (10%) currently has simulation tools. important for agent performance analytic systems. Noted features of dashboard systems included real-time performance information shown to agents, the flexibility in creating the dashboard, and the ability to see information at a glance. Consistency and another voice in the hiring process were favorite features in hiring and recruiting tools. Systems Next in Purchasing Plans Over one-quarter (26%) of the survey participants noted that agent performance analytics would be next in their purchasing plans. Twenty-two percent plan to purchase dashboard reporting, while 12% are looking to buy an automated customer satisfaction measurement tool. Eleven percent plan to purchase a quality monitoring system, and 10% plan to buy simulation tools, e-learning systems, and workforce management software. Hiring and recruiting tools are in the purchasing plans of 7% of respondents. Systems Provided by Same Vendor When asked if the current technology in the center was provided by the same vendor, over three-quarters (76%) said each technology was acquired from separate sources. Twentyone percent said two or three systems came from the same vendor, but only 3% had three or more from one source. Favorite Features of Systems The survey participants were asked for favorite features of each system. Most noted for workforce management tools were the intraday forecasting and performance capabilities, as well as real-time adherence modules and ease of use/maintenance/administration. For quality monitoring systems, the respondents most often mentioned the ability to capture both screen and voice and also monitor all calls coming into the center. The ability to use e-learning systems during idle time and finish training online were the top choices for features in that product, while compliance reporting and daily performance reports were Benefit of Acquiring Technology from Same Vendor The largest percentage of survey respondents (29%) felt that simpler implementation was the biggest benefit in acquiring technology from the same vendor, while 19% noted that minimal data entry would be a big benefit. Another nineteen percent felt having only one vendor to manage would be helpful, while 17% saw no benefit in buying from one vendor. Another 12% noted having less infrastructure as a selling point. Drawback to Acquiring Technology from Existing Vendor Thirty-four percent of the participants felt that acquiring additional tools from an existing vendor would sacrifice best of class capabilities offered by others. Thirty-one percent worried Continued on page 14 2

3 W F M S u r v e y SWPP conducts a survey each quarter on critical workforce planning topics. These results will be published in upcoming issues of On Target, as well as on the SWPP website in the members-only Library section. You may fax this page to or fill in the survey online at This quarter s survey focuses on agent adherence. Responses to the first two questions will allow us to segment the answers by size and type of call center to contrast and compare workforce management practices. Survey results will be completely anonymous. Responses to the first question will allow us to segment the answers by size of call center to contrast and compare workforce management practices. Survey results will be completely anonymous. 1. What is the size of your operation in total agent seats at all locations? Under Over What industry do you represent? Telecommunications Travel Financial Insurance Utility Retail/Catalog Health Care Outsourcer Other: 3. What is the schedule adherence goal in your center? <90% 90% 91% 92% 93% 94% 95% 96% 97% 98% 99% Other: 4. What is your average actual adherence? <90% 90% 91% 92% 93% Respond and Win! 94% 95% 96% 97% 98% 99% Other: 5. Do you use an automated real-time adherence monitoring tool? Yes No If yes, what adherence statistics do you measure with this tool? 6. Is adherence part of an agent s performance score? Yes No If yes, How important is adherence in the score (or what percentage)? 7. Does adherence play a role in agent ranking for schedule selection? Yes No 8. What incentive programs do you have in place to encourage adherence? 9. How are adherence statistics communicated? Choose all that are applicable. Directly to the agent To the supervisor To the agent and the supervisor Posted for public display Other: 10. What is the biggest cause of non-adherence in your center? Not only will you receive a report of our findings, but you ll have a chance to win a free SWPP membership for responding to the survey. Please return to SWPP by September 30, Congratulations to Theresa Shripka of ING Direct, who won a free SWPP membership last quarter for completing the SWPP survey. Name: Company: address: Agent adherence 3

4 B e g i n n e r s G u i d e Three Steps for Attacking Adherence Problems By Penny Reynolds, The Call Center School Ask a group of workforce management professionals about their biggest problem and a high percentage will tell you that it is schedule adherence. Enormous effort goes into forecasting workload, calculating staff requirements, and creating staff schedules. But all that hard work goes down the drain when the frontline staff don t stick to the schedule plan. This lack of schedule adherence is frustrating and it s expensive. (See Managing by the Numbers article, page 9.) So what can be done about it? This article will outline some of the reasons that agents may not be adhering to their planned schedules and explore some potential solutions to the problem. For all you workforce planners out there, it s time to step away from erlang and workload distributions and enter the world of psychology and behavior analytics. Performance Management Approach Figuring out why people behave a certain way is called behavior analytics and the application of this psychological science in the workplace is called performance management. A structured performance management model is made up of the following steps. 1. Define performance standards and objectives. 2. Measure existing performance. 3. Identify gaps in what we want versus what we have. 4. Determine root cause(s) for the performance gap(s). 5. Apply a behavioral solution to address the underlying cause. Let s take a look at this model as it applies to schedule adherence. The first step is to define performance standards and communicate them to the staff. This involves defining the precise times that a person is expected to be on the phones, take breaks, go to lunch, work on other activities, and so on. This needs to be defined and then communicated clearly as the expectation. Don t assume that when you specify a start time of 7:30am that it s perfectly clear what that means. Some staff might take that to mean in the parking lot or in the front door at 7:30, clocked in at the time clock and in the break room getting coffee at 7:30, or even at their desk getting organized at 7:30. If you mean logged in at 7:30 ready to take a call, then be explicit. Another part of this definition will involve the leeway that exists in meeting the exact numbers. What percent time out of adherence will be allowed at a maximum? Are there various levels of adherence that will earn them a better grade than others? And what are the rewards and/or consequences for meeting the goal or not meeting it? All this needs to be precisely defined for everyone in the center and communicated clearly and often. Part of the communication about schedule adherence needs to be education about the numbers. Be sure the staff understand the relevance of adherence and why it s so important that each person be in his/her seat on time. Every member of the team should understand the power of one when it comes to call center staffing and know the impact they make on speed of service, occupancy, and bottom-line cost. (See What Difference Does One Person Make? article in Summer 2003 issue of On Target.) The next step is to measure performance. Unlike some other qualitative measures of performance, schedule adherence lends itself to being measured quite easily. Note the login/logout times compared to schedule and note the deviations, both in terms of total minutes as well as a percentage of total hours scheduled. If the deviation meets your defined grace period, then there s no performance gap. However, if the deviation from schedule is more than allowed, the next step is to identify the reason that person is not meeting the expectation. Three Reasons for Non-Performance There are three basic reasons why an employee doesn t do what is expected. These reasons are: 1. Don t Know 2. Can t 3. Won t 4 Let s take a look at these from a schedule adherence perspective. First, the agent may not know what is expected in terms of schedule adherence. Have expectations of start/stop times, breaks, and off-phone time been communicated clearly? Does the agent know how much deviation is allowed and what the consequences will be for adhering or not adhering? Make sure each person understands the schedule contract, grades of adherence, and consequences for following or not following the plan. The other instance of don t know is when agents are fully aware of the expectation, but are not getting enough feedback about how they re adhering to the work plan. Make individual schedule adherence statistics available to them on a regular basis, preferably daily. If the two don t know issues have been covered (they re aware of the expectation and have been receiving their adherence numbers on a regular basis), then the next possibility for cause of the problem is in the can t category. Perhaps they don t have the knowledge/skill to perform, or there is a barrier or obstacle preventing proper performance. While the can t reason can explain some other performance issues in the call center, it is generally not an underlying cause of adherence problems. Most agents are perfectly capable of following their work schedules. Occasionally there is a barrier to adherence (long calls forcing them to vary from scheduled stop times or severe understaffing that make it difficult to actually break away from phones when planned). However, most schedule adherence issues do not fall into the can t category. This brings us to the final possibility the won t category. This is the reason for most adherence problems. Most employees with an adherence problem have willfully decided not to stick to the schedule. The reason for this behavior is most often the lack of a proper behavioral consequence. If your call center has no real consequences associated with schedule adherence, then think about what Continued on page 5

5 B e g i n n e r s G u i d e Continued from page 4 happens to two agents seated side by side. Sarah is the picture of punctuality while her neighbor Alex is consistently late. With no formal system of consequences in place, Sarah will actually be getting negative consequences by being on time she has to handle more calls and workload is heavier because someone is missing. On the other hand, Alex is getting positive consequences by his behavior he may be getting another few minutes sleep in the morning or be spending extra time in the coffee area and he doesn t have to take as many calls. Unless a regular, predictable system of consequences is in place, you may be rewarding the behaviors you want to disappear, while there s nothing to support the actions you want to see. Applying consequences is critical to shaping behavior. While some consequences occur naturally, others will need to be consciously applied. For example, agents may demonstrate some desirable behaviors such as telephone etiquette because they get an immediate consequence for doing so a friendly, thankful reaction from a customer. However, other behaviors like schedule adherence don t come with natural built-in positive consequences, so you will have to apply consequences to shape the behavior you want to see. Both positive and negative consequences can be used to influence behavior. If you can catch someone demonstrating desirable behaviors, then you will want to immediately apply a positive consequence. Positive consequences work more effectively than negative consequences. People like positive reinforcement better, it produces a less stressful environment, and it also maximizes performance. Negative consequences can also be used, but they are not as effective as positive consequences. Negative reinforcement will shape performance, but it generally produces behaviors that are just enough to get by. Discipline and punishment will generally stop a negative behavior, but in some cases only for a little while, and it will never encourage employees to give maximum effort like positive consequences can do. Sometimes situations occur where even though positive consequences are given for desired behaviors and negative consequences are attached to the wrong behaviors, employees still do the wrong things. This is because there are other characteristics of consequences in addition to just the positive and negative aspects. In addition to the positive/negative aspects associated with a consequence, there are also the aspects of how personal a consequence is, how immediate it is, and how certain it is. These other aspects of a consequence can sometimes outweigh the positive/negative aspects. Let s look at an example in the table below where positive aspects for adherence and negative aspects for nonadherence are in place, but still don t yield the desired result. Examine both the positive and negative effects associated with this lack of schedule adherence, along with the personal, immediacy, and certainty aspects of the consequences. Even though there are many negative consequences associated with a lack of schedule adherence, the employee may continue to do it. Two of the negative consequences are of benefit to the overall call center and customers, but not felt as a personal effect. The bad appraisal and loss of bonus are also negative, but they are not immediate. Those things will likely happen sometime out in the future, and may be viewed as uncertain by the employee. These negative consequences may be outweighed by the positive consequences. The benefits may include an extra ten minutes of snooze time in the morning, or an extra few minutes to socialize in the break room, not to mention fewer calls to take. All these consequences are personally felt by the employee, and they re all immediate and certain. Even though they re not as 5 significant as the negative ones, the fact that they re personal, immediate, and certain may sway the employee to continue his errant schedule behavior. The key when developing a plan of consequences is to apply consequences that are positive to shape desired behavior. However, it s not enough that the consequence is positive. It also has to be personal (something that means something to the employee), immediate, and certain for it to work as an influence on behavior. With this in mind, think about ways to make the positive and negative consequences more immediate. A supervisor that is waiting at the agent s workstation with a warning note when he comes back from break may send a stronger message than simply reporting adherence numbers at the end of the week along with a warning. Some call centers choose to project the real-time adherence screen up for all to see, so that other employees can apply some peer pressure on the spot for agents coming back late from break. Conclusion The job of the workforce planner isn t over when the schedules are complete. Making the plan a success involves working with supervisors and frontline staff to ensure that everyone is where they re supposed to be. Schedule adherence will be much higher when the frontline staff have been educated on the relevance and importance of sticking to the plan and provided with regular feedback on how they re doing. Couple this with a system of appropriate rewards and consequences and you ll see schedule adherence steadily and surely improve. Penny Reynolds is a Founding Partner of The Call Center School, a company that provides a wide range of educational offerings for call center professionals. Penny is a popular industry speaker and is the author of numerous call center management books, including Call Center Staffing: The Complete, Practical Guide to Workforce Management and Call Center Supervision: The Complete Guide for Managing Frontline Staff. She can be reached at or by at: penny.reynolds@thecallcenterschool.com.

6 A d v a n c e d T o p i c s It s Budget Time Do You Know Where Your Shrinkage Is? By Jennifer Behrens, Thrivent Financial for Lutherans Heading into a new budget cycle is a great time to be prepared with what the coming year s anticipated shrinkage levels will be. What shrinkage number will you use? Shrinkage is a critical input in the staffing model and deserves a thoughtful approach. Determining shrinkage is partly a reflection on experience and partly a look into the future. For the purpose of this article, shrinkage is defined as the percentage of an agent s paid time when the call center is open and the agent is not available to handle calls. Whatever your definition, make sure the call center leadership and workforce management understand it and agree upon it. A Proactive Approach Since next year is likely to be different than this year, determining next year s shrinkage should be more of a planning exercise than an analysis of past data. Consider employing a worksheet to facilitate the shrinkage planning for the year ahead. Completing the worksheet can be a collaborative effort between workforce management and the supervisors, building agreement and understanding throughout the process. Ask the workforce management staff to complete the fixed portion and have the supervisors complete the variable section. Shrinkage Calculation Worksheet Start by calculating the total paid hours projected for the year ahead. You may choose to subtract out holiday hours when your company is closed, using the rationale that those hours don t need to be included since there is no workload and no one is staffed. If your center routinely uses overtime, you may want to include projected overtime hours in the total. Once you ve determined what paid hours to include and exclude, the final total paid hours number will be your denominator as you complete the worksheet. The worksheet assumption may be that as attrition occurs, agents are replaced by people with similar off phone needs. The body of the worksheet has two sections: fixed shrinkage and variable shrinkage. Fixed Shrinkage The fixed shrinkage elements are time off phone events that are benefit related including vacation time, sick time, personal time off (PTO), floating holidays, paid breaks, FMLA, funeral leave, jury duty, and other benefits your organization may offer. How are they calculated? PTO/vacation/sick hours are calculated based upon each agent s fulltime or part-time status and his/her years of service. If your agents carry time over from year to year, consider including those hours, too. 6 Floating holidays can change from year to year. January 1st falls on weekends in 2005 and again in Are you open for business on January 3rd and getting an extra floating holiday as a result? Include employee floating holiday hours based upon their full-time or parttime status depending upon your company s policy. Paid breaks take a little finessing to calculate. We know that shrinkage for two 15-minute breaks is 6.25% for a typical 8 x 5, 40 hour work week. How many part-time agents take only one break during their shift? How many alternative full-time schedules are worked in your center? Two 15-minute breaks for a 10 x 4 schedule is 5% shrinkage. Look at your present break shrinkage and your schedule mix. If you expect to have a similar mix of schedules in the future, using today s break shrinkage number may be acceptable. If your center is driving toward more or less Continued on page 7

7 A d v a n c e d T o p i c s Continued from page 6 alternative and part-time schedules, adjust the percentage accordingly. FMLA, funeral leave, and jury duty are likely to be estimates only. Using the data you ve been collecting for these elements, determine if these activities are likely to remain fairly constant. If you are seeing a positive or negative trend in these areas, then make the appropriate adjustments. We ve added a few requirementrelated events to our fixed shrinkage because call center leadership doesn t have the discretion to eliminate or target these events for reduction. These may be regulatory, corporate, or department requirements. Determine per person hourly requirements as the worksheet input. Variable Shrinkage Variable shrinkage includes the elements driven by call center leadership that call center supervisors have some control over. Some likely elements are training, meetings, development, projects, and unproductive time. Create a list that applies to your environment. Ask the supervisors to consider how many hours they plan to allow for each activity. Encourage them to talk with their partners in human resources, corporate event planning, product services, training, and IT to learn about product roll-outs that will require training, changes to benefits with corporate-wide meetings planned, or a new version of a software application requiring training, etc. The supervisors can t be expected to hit the nail on the head with these hours, but to provide estimations that are logical and easily explained to leadership. Unproductive time is an element that deserves a closer look. Unproductive time for shrinkage is often miscalculated as 100% minus the center s adherence average. If a center is achieving 95% adherence, then unproductive time is 5%, right? There s a bit of a twist to consider. Adherence measures the time the agents were scheduled for phones. Remember we re calculating what percentage of their time will be spent off phones. Here s a sample calculation for a center with 95% adherence to schedule, 10,000 login hours, and 13,500 total paid hours:.05 x 10,000 = total unproductive hours /13,500 =.037 or 3.7% unproductive shrinkage for worksheet Since the inputs require login hours, come back and calculate this item last on the worksheet, after totaling hours for all off phone activities and subtracting that total from total paid hours to determine the login hours. When the worksheets are complete, bring the groups together and look for things that may have been overlooked. Once finalized, agree as a group to the estimations. There, you ve calculated your shrinkage, and you re done. Not so fast you ve got to share the results up the line. Selling the Results Now that you ve done your homework and there s a solid partnership and understanding among workforce management and the supervisors, it s time to present the worksheets to call center leadership. They accept the number right off the bat because they know how meticulously it was calculated, right? Your experience may be closer to That s a higher percentage than we were expecting, followed quickly by How does that compare to the industry average? and Have you benchmarked with companies in our industry? Benchmarking Caution What does it matter if your shrinkage is higher or lower than another center? The comparison tells you nothing. Have you ever found another center exactly like yours? How do your company s vacation benefits compare? Do your products change more frequently than theirs requiring more training? Do you include overtime in your calculation for total paid hours and they don t? There are so many variables that make up the shrinkage number that you are unlikely to find a relevant comparison. Industry Average If your leadership believes the industry average to be between 25% to 35%, and you present a higher percentage, refocus leadership internally by 7 sharing the logic behind the hours in each element. Being outside the average is OK because your shrinkage percentage applies specifically to your unique situation. Reducing Shrinkage If reducing shrinkage is the directive from leadership, the worksheet provides you with a tool to begin that process. Go back to your supervisors and call center leadership with the worksheet. Ask them specifically which elements they want to target for reduction. And, most importantly, define a plan to drive the reduction. Simply saying, We ll hold fewer meetings isn t enough. Which meetings will continue to be held and which specific meetings will not be scheduled? Without the meetings, how will the messages be communicated? Demand specifics or the plan won t work, resulting in understaffing. The results of the shrinkage reduction must be realized before the new shrinkage level should be used in staffing models. So, if leadership is targeting a lower shrinkage to be used for next year s staffing models, the reduction plans must be proven effective prior to the end of this year. The analysis isn t complete once the shrinkage level is accepted. Review the shrinkage you actually experience against the model you created. Are you trending close to the projections? What were you surprised with that increased shrinkage unexpectedly? Make a note to include the event in next year s model if it s likely to occur again. Shrinkage is likely to change year-toyear in our dynamic environments. The most important shrinkage management tactics are keeping up with the changes through proactive reviews with a focus on your unique environment and communicating openly with leadership. Now, bring on the budget. Jennifer Behrens is Advanced Business Process Analyst, Customer Interaction & Investment Services, for Thrivent Financial for Lutherans. She may be reached at Jennifer.behrens@thrivent.com.

8 Excel Tips for Workforce Planning Professionals The OFFSET Function Thank You to By Mike Andrews, Workforce Management, Cincinnati Bell Sponsors Platinum Level Figure 1 Gold Level The OFFSET Function can be used in a number of ways to define a specific cell or a range in a worksheet that is/are a specified number of rows and columns from the reference cell. You can use the OFFSET Function to perform lookup operations or to locate information/data that is not in a fixed location or to dynamically update charts. By combining the OFFSET Function with other functions such as IF, MATCH and COUNTA, you can turn this into another power tool to use in your workforce planning efforts. This article will show you how to use the OFFSET Function to dynamically graph annual call information data. In this example, our requirement is to separately graph calls offered information for 2003 and The syntax for the OFFSET function is OFFSET(Reference cell, rows moved, columns moved). In Figure 1 (Cell D5), the formula =IF($G$4=2004,OFFSET(C5,12,0),C5) says no more than if you type 2004 in cell G4, the rows will be OFFSET by 12 and the columns by 0, otherwise the data set will revert to the 2003 range (cells C5 through C16). Be sure to autofill cells D6:D16 to complete your data range. At this point, type 2003 in cell G4 and notice the change to the graph. Building the Chart To build your chart, highlight cells B5 through B16. Use your Control Key and highlight cells D5 through D16. Run your Chart Wizard for either a line or column chart and you re all set. As discussed previously, when you change the year in cell G4, your chart will change with the appropriate data. Our goal as always is to provide you with information that may be applied to your workforce planning efforts. Be sure to download the corresponding spreadsheet on the SWPP website. It contains another example of how the OFFSET Function could be used. Again, the web is full of examples that could be modified to meet your needs. Mike Andrews, a certified Microsoft User Specialist in Excel, is the Team Leader, Workforce Management at Cincinnati Bell. He may be reached at mike.andrews@cinbell.com. 8 Silver Level Bronze Level Interested in becoming a sponsor? Call Vicki Herrell at

9 Ask the Workforce Wizard What are Long Abandons? Question: I heard a new term today that I had never heard before long abandon. The user of the term stated that at a previous job, they only counted those abandons which occurred after the service level target. For example, if the service level goal was 80% of calls answered in 60 seconds, the company excluded all calls that abandoned in less than 60 seconds from their reporting. Our company excludes short calls, which are defined as calls that abandon in less than 10 seconds. I have seen this practice at several companies with the intent being to exclude hang ups for reasons beyond the control of the company (e.g., customer discovers they no longer need service, have to hang up to take another call, etc.). Can you please tell me if most companies exclude calls for purposes of calculating NCO, service level, or abandons? Also, have you heard of or used the term long abandons? Any help on this would be greatly appreciated. Answer: We were quite intrigued by the concept of long abandons, and thought we would ask some other real-life experts how they handle abandons. Here are some of the responses: I ve definitely heard of short abandons (and seen them defined as 10 seconds or even six seconds), but I ve never seen the term long abandons. If one factors out all the abandons prior to the service level target, I would think that could mask potential problems that I d certainly want to see. We calculate both Total Calls Handled and Calls Handled > than five seconds (this is to account for the hang-ups, etc). For service level, we have call centers in a couple of areas that require a longer recording (to identify that the call may be recorded or other required recording by state law), so for calculating service level in those centers, we add the amount of time the extra recording takes. So if our service level goal is 80/20 and the recording in that center adds 8 seconds, then the service level for these centers is calculated at 80/28 since the other centers do not require the recording. That keeps our metrics consistent and equitable. I ve never heard of long abandons, or of calculating the abandon rate in the manner you described our abandon rate would be near 0% if we calculated that way. I suppose if you have a captive audience of callers that would be okay, but for those of us in a sales environment, a call answered is a potential sale so we want to miss as few as possible. If I reported the abandon rate that way I d have a hard time explaining to our executives that we have an almost non-existent abandon rate, but need to have a wide margin between NCO and NCH for budgeting. I d also be concerned that I was missing key data needed to appropriately staff to improve our calls handled and/or service level not to mention our sales conversion. We omit calls that abandon under nine seconds from abandon and service level calculations. I am not sure that this is the most accurate way to portray the actual day s operation, but that is our current standard. The rationale used is that a) these are callers who have misdialed or b) we could never attain that Service Level/ASA so why penalize ourselves? Based on the way I understand our ACD system, the call is not counted as abandoned until it is actually in queue, so the caller has already heard the greeting and know if they have misdialed or not. So the first reason is not really valid. The second reason is probably open for debate, and maybe we should not penalize the operations people for caller behavior. I can sort of buy-in to the logic that long abandons shouldn t impact the percentage of abandonment so long as they still impact service level. You d argue that since we have Y seconds to answer a call, why hurt our abandonment percentage for not being able to get to a call that abandons so quickly. On the other hand, I like my measures simple, and prefer no forgiveness. An abandoned call is an abandoned call. In a previous call center life, we forgave abandons of two seconds or less. My experience with that is that our abandon percentage was almost non-existent when we had 80%+ service levels, but once service level dipped into the 60s and 70s, we d see abandon percentage increase. I just wasn t a fan of masking our results and having two different denominators for abandon percentage and service level. Have a tough question? Send it to wizard@swpp.org and we ll try to find an answer! 9

10 E v e n t s C a l e n d a r Tradeshows/Conferences/SWPP Meets: Name Date Location 2005 Aspect Professional Development and Contact Center Road Show Sept. 7 Dallas, TX 2005 Aspect Professional Development and Contact Center Road Show Sept. 8 Austin, TX Annual Call Center Exhibition (ACCE) Sept Washington State Convention & Trade Center SWPP Meet Sept. 19 Seattle, WA 2005 Aspect Professional Development and Contact Center Road Show Sept. 22 Simi Valley, CA ICCM Sept Bellagio Hotel SWPP Meet Sept. 27 Las Vegas, NV 2005 Aspect Professional Development and Contact Center Road Show Oct. 6 Seattle, WA 2005 Aspect Professional Development and Contact Center Road Show Oct. 11 Chicago, IL 2005 Aspect Professional Development and Contact Center Road Show Oct. 12 Toronto, Ontario ebiz the Aspect User Conference Oct Hilton Riverside New Orleans SWPP Meet Oct. 16 New Orleans, LA IEX EMEA User Conference Oct Queen Hotel Chester, UK Education: Title Date Presented By Time Forecasting Fundamentals Sept. 8 The Call Center School 2:00 ET Introduction to Workforce Management Sept. 8 The Call Center School 11:00 ET Call Center Math Sept. 15 The Call Center School 11:00 ET Calculating Call Center Staff Sept. 15 The Call Center School 2:00 ET Scheduling Principles & Problems Sept. 22 The Call Center School 2:00 ET Managing Daily Staffing and Service Sept. 29 The Call Center School 2:00 ET Attendance and Adherence Strategies Oct. 6 The Call Center School 2:00 ET Making the Most of WFM Oct. 13 The Call Center School 2:00 ET Call Center Consolidation Oct. 20 The Call Center School 2:00 ET Workforce Management Oct The Call Center School 8:00 a.m.-5:00 p.m. ET Two-day classroom seminar Atlanta, GA 10

11 8: %20 Quantifying the Impact of Schedule Adherence It s just a few minutes a day. What s the big deal? Can you quantify the impact that a lack of schedule adherence has on your call center? This article will offer several approaches for putting some numbers to your schedule adherence or lack of it. Impact on Service, Occupancy, and Cost The table below shows what happens when we have 350 calls per hour with an average handle time of 320 seconds (or 62 erlangs of workload). Let s assume we had planned to have 69 bodies in chairs to handle those calls in order to deliver an ASA of less than 20 seconds. What happens if just 5% of our staff (in this case, just 4 people) are not adhering to their work schedule during this period? Note what happens with 65 staff in place. The first impact is the obvious degradation of service, dropping from a 14 second ASA to a 73 second ASA, an extra 59 seconds of delay time per call. With 69 staff in place the occupancy would be a reasonable level of 89%, but when 4 people drop out, occupancy shoots up to 95%, a level that will be painful for those people that are actually on the phones. And finally, with an extra minute of time on each call due to queue time, we ll be paying the phone company almost $35 per hour extra for the time our callers are waiting. This situation is worsened by the fact that as delays increase and occupancy rises, the talk time and after-call work time will likely go up, causing a higher workload, and an even worse speed of answer, occupancy, and cost increase than what we ve noted here. Therefore, you can view the impact of non-adherence as these three detrimental effects, or you can simply plan for the loss in the shrinkage calculation and pad in the needed extra staff. Either way, there s a significant cost to the center. Other Approaches Another way to look at the impact of non-adherence is to calculate the lost time cost. Assume that there is a loss of 15 minutes per day per agent in a call center that houses 100 staff. This 15-minute loss per person adds up to about 65 hours a year of lost time. At an average cost of only $10 per hour, that s over $650 of lost time per person, or $65,000 for the entire center. Simply vary the lost minutes per day, the wage rate, and the number of seats in the call center to arrive at what this loss means to your center. Other centers may choose to look at the salary cost of doing business for a day. With 100 agents making $12 per hour and scheduled to be on the phones seven hours per day, that s a salary cost of $8,400 for the day. If your adherence is only at 90%, then lost time of 10% means a cost of $840 for the day where people are being paid to be on the phones but not available. Translate this daily number into monthly and/or annual numbers for a overall cost. Is it something you can afford? Conclusion Regardless of how you add up the numbers, the end result is the same poor adherence costs you money. Quantifying the loss may help you communicate the importance of adherence to frontline staff and supervisors. It can also help justify the tools needed to better track and report it, as well as fund reward programs to provide needed incentives to improve overall adherence. Sponsor a Workforce Planning Awareness Day. To increase agent understanding of workforce management, our team sponsored a Workforce Planning Awareness Day. It was a fun event, and it really helped the agents understand what we do and why. We set up a display in a common area with information about workforce planning (WFP), which included Q&A brochures, WFP word find, posters, etc all created by the WFP team and a game station. The agents could visit the display booth and game station on their breaks, lunches, or before or after their shifts. We also passed out Krispy Kreme doughnuts and sent out s throughout the day with trivia questions and awarded prizes to the first responder with the correct answer. It was a great day, and the agents had fun while learning more about the purpose and processes of workforce management. Note: This tip provided by SWPP Board Member Kristi Comfort of Assurant Solutions. She may be reached at Kristi.comfort@assurant.com. 11

12 W F M S u c c e s s S t o r y GE Consumer Finance Reaps Benefits of TotalView Solution GE Consumer Finance, a unit of General Electric Company, with $77 billion in assets, is a leading provider of credit services to consumers, retailers and auto dealers in 35 countries around the world. GE Consumer Finance offers a range of financial products, including: private label credit cards, personal loans, bank cards, auto loans and leases, mortgages, corporate travel and purchasing cards, debt consolidation, and home equity loans as well as credit insurance. In 1995, GE acquired the credit card business of Coles Myer one of the largest non-bank credit card operations in Australia. Today, GE Consumer Finance Australia (GECFA) has contact centers in Burnley, Paramatta, Brisbane, Auckland, and India. These centers employ around 900 agents across 23 sites. Opening hours vary, depending on the site, but GECFA contact centers are generally operational from 7 a.m. until midnight, 7 days a week. On average, the centers receive 7 million calls per year in addition to 300,000 faxes and 1.2 million applications via their Web-based system called e-class. The Challenge Following GE s acquisition of AGC, it was determined that substantial efficiencies could be achieved in the area of workforce management. The legacy system, one of the top three workforce management systems on the market, was deemed incapable of meeting GECFA s requirements. This resulted in GE doing a thorough evaluation of a number of workforce management solutions before choosing the IEX TotalView Workforce Management system. In the year since installation, GECFA has realized significant cost savings as well as improved accuracy of forecasts and rosters. The functionality TotalView provides has led to a greater appreciation among all levels of the business about the value of a defined workforce management process. The GECFA s contact center for Cards implemented Skill Scheduling to cope with the seven different skills and complexities of the area. The system also boosted confidence in forecasting the skills required to cope with different campaigns and agent knowledge. The flexibility of TotalView was highlighted when Coles Myer launched their Source Mastercard. GECFA was able to proactively introduce this product line with an understanding of the number of people required to handle the influx of calls. An additional 300 temporary agents were scheduled using TotalView to achieve GECFA s targets. Real Benefits that can be measured and seen TotalView s Real Time Adherence module has been credited with providing one of the biggest benefits for the organization. Adherence has improved by 30 percent since installation of TotalView to 95 percent. Monthly reports are sent to senior managers so that they can maintain visibility into the centers performance. The availability of intraday tools to view center performance and manage agent schedules is another of the top benefits GECFA has identified in using TotalView. At a glance, managers can get a clear picture of how the center has performed so far and see the outlook for the rest of the day. The visual representation of agent schedules, including off phone activities, gives managers the ability to make informed decisions about whether or not to take staff away from scheduled activity. Back office activities are also scheduled using TotalView, with a target service level, of all faxes being answered within 30 minutes. Agent WebStation has proved to be a huge success with managers and agents alike. Agents are now able to view their schedules and swap shifts, which has resulted in decreased sick leave and improved morale. Users can customize their screens by setting their own meeting messages, creating a more personalized interaction with TotalView. WebStation has also empowered Team Leaders to manage schedules and intraday activities, allowing them to use their time more efficiently. The workforce management team is able to utilize the What if functionality of TotalView prior to implementing a change in the length of paid breaks, to model and report the impact of this change to key stakeholders. The increase in the paid break allowance has now been implemented, and the cost of the extended breaks has more than been absorbed through improved adherence to schedule. GECFA has reaped the benefits of TotalView in a short period of time and they have only just scratched the surface of the functionality available in this solution. IEX TotalView has delivered all the benefits we planned it would. It was great to have a workforce management tool live up to expectations and deliver real savings to our business, said Stuart Beaumont, planning & resource management leader for GECFA. Mark Your Calendars! 2006 SWPP Annual Conference March 15-17, 2006 Opryland Hotel Nashville, TN 12

13 Does Accuracy Matter? Ric Kosiba, Ph.D., President, Bay Bridge Decision Technologies Our Struggle With Accuracy I am very proud of my company, Bay Bridge Decision Technologies. We ve brought together a very talented group of mathematicians and industrial engineers who spend a good chunk of their day analyzing ACD data, developing hiring optimization algorithms and contact center simulation models, and ensuring simulation model accuracy against real world call center data. When we started Bay Bridge, we built our first prototype CenterBridge system using Erlang equations to develop staff and budget plans. Everyone in the call center industry used some variant of the Erlang equation to do staffing analysis, so we thought we would as well. We plugged the results of our Erlang model and compared it to real-world ACD data and found that the results were not very close. So, we went back to the drawing board. We researched and developed modified Erlang equations for use in CenterBridge. This time we had more luck. We found that we could get more accurate for a particular call center (although still not as accurate as we d like), but that model did not translate well to other call center data. In other words, for our models to be closer to accurate, using a modified Erlang equation, we had to build a different model for every call type. There was a ton of math involved in doing this, a lot of time, and the results were still not as close as we d like. But this makes a lot of intuitive sense. Every call type and call center customer is not the same. We all know that analyzing a technical support contact center is different from analyzing a sales center. We all know this, yet most of us use the same equation to determine staffing across all of our different call types. We had to go back to the drawing board again. We found that we could get amazing accuracy with a well known, but slow, technology called discrete-event simulation modeling. So, our engineers went to work on speeding up the simulation model runs. When they finally succeeded, we had our first version of CenterBridge. The Big Question We spend an awful lot of time building these custom simulation models for our customers. To get this accuracy, we have had to hire expensive engineers, invest our time and resources into software and processes designed to ensure speed and accuracy, and go to great lengths to analyze our customer s ACD data. What a pain. I cannot tell you how many times we have asked ourselves: Does accuracy really matter? It would sure make our lives easier if accuracy was not so important. What does accuracy mean to your staffing numbers? It is common knowledge that Erlangbased equations tend to overstaff. The real problem most analysts face is that they do not know, for their centers, how much Erlang tends to overstate the staffing requirements. During our normal course of business analyzing ACD data, building discrete-event simulation models, and comparing models to reality we also take the step of comparing models to Erlang equations. Here s the rub. For the same center, Erlang may overstate requirements anywhere between 2% and 7%. It is not always two, and it is not always seven, it definitely fluctuates. So you cannot hide the inaccuracy with fudge factors or schedule inflexibility factors, because the fudges would move around. Figure 1 represents a head to head comparison between Average Speed of Figure 1: A comparison between Erlang, Simulation, and Actual center performance. 13 Answer predictions. It compares what actually occurred at a contact center, with what was simulated using Discrete- Event Simulation, versus an Erlang prediction. As you can see, Erlang always overstates the requirement for this contact center (and from our experience, every contact center). What does accuracy mean to your costs? Clearly, being off anywhere from 2-7% in your staffing figures translates to large inaccuracies in your budget figures. While we can sometimes tolerate staffing inaccuracies as a safety net, it certainly does not go over well with our friends over in finance. Figure 2 represents the trade-offs between staffing accuracy and cost per call. It is important to know this trade-off for every one of your centers. Figure 2: Trade-off between cost per call and staffing forecast accuracy. But I would argue that there is another cost that may be more problematic. The extra shrinkage or lost time that becomes available when there is invisible, Erlang-derived, slop in your contact center network, breeds sloppiness in your center. Any time that you are consistently overstaffed, it means your center management will be able to slack off to some extent. Those inaccurate 7% weeks are not conducive to running a tight ship. What does accuracy mean to what-if analysis? Throughout the planning cycle, what-if analysis is very important. These what-ifs are often used to make important strategic decisions, such as: Continued on page 14

14 Continued from page 13 Should I close a center? Should I offer this cross-sell program? Should I combine agent groups? Should I outsource? How much of my business should be outsourced? Clearly, overstating your staffing requirements, may possibly lead you down the wrong path on a number of these questions. Looking back on Figure 2, ask yourself, If I overstate my staffing requirements, and put 8% additional cost into my forecast, will we be more or less likely to outsource my center? These are big, big, strategic decisions. Short Term is Different From Long Term Now I know that for tactical workforce management, this issue is very much lessened. Certainly, being off a few percent for the next 15 minutes does not pose the same risks as being off in your budget and long-term staff plans. Most workforce managers I speak to have their own fudge factors, kept in the back of their heads, to help them manage this inaccuracy on a day-of basis. This intuition is one of the differentiators between an experienced workforce manager and a newbie. Does My Planning System Use an Erlang-Based Model or Discrete-Event Simulation? Simulation is one of the words that many folks can use to describe many methods of analysis. Indeed, an Erlang equation is itself, a form of simulation (impress your workforce management friends Erlang is a closed-form equation, or a continuous simulation ). The type of simulation model that leads to the accuracy described in Figure 1 is called Discrete-Event simulation. What makes this form of simulation so powerful is that it can, and should, take into account behaviors of your customers, such as their patience on the phone and their handle times. You can create virtual, computerized customers in your Discrete-Event Simulation, that have the same statistical attributes that your real customers have. By getting these behaviors right, you can get the accuracy that we like to brag about. Most of us use spreadsheets to do our strategic planning. A simple way to tell if you are using Discrete-Event Simulation or some Erlang derived equation is to check whether your tool outputs all the standard metrics: service level, average speed of answer, abandons, and occupancy. If your model does not explicitly and accurately display your abandon rates, it is most likely an Erlang equation and not Discrete-Event Simulation. Does your model first develop requirements? Discrete-Event Simulation works the other way it presents to you the service expected given your handle times, staff levels, and call volumes. Erlang-based equations usually display requirements, given your service level goal, first. Usually, you ll also have to provide to your spreadsheet some sort of schedule inflexibility factor to help fudge your Erlang equation. The Importance of Model Validation No matter what type of tool you use to help you with your strategic planning, it is important that you validate the accuracy of your tool. To do this, simply plug into your planning tool, say, for last week s historical center performance, the number of staff you had, the call volume you received, and the handle time you saw. If the resulting service level, speed of answer, abandon level, and occupancy of your historical center performance doesn t come very close to what your model says you should have achieved, then you have an accuracy issue. Now, the most significant area that accuracy will help you is probably bad decision avoidance. When considering big changes, like outsourcing, or opening or closing centers, you want to make sure your analysis is spot on. But even in the course of putting together plain vanilla budget plans, being able to squeeze out five percent of your staffing costs is pretty significant. It could help save your company money. Accuracy matters. A lot. Ric is a charter member of SWPP. Feel free to reach him at EDK@baybridgetech.com or (410) Workforce Management Survey Results Continued from page 2 that changes in one application could affect others, and 23% did not want to be locked into a single vendor. 14 Conclusion There is a lot of press and vendor effort going into workforce optimization (WFO). However, this survey gives the impression that the call center community has not quite embraced the concept yet. Many centers have acquired quite a bit of technology, but are not purchasing from one vendor. And while they see benefits to one-stop shopping, they are not ready to give up best of class capabilities from other vendors. Since we are early in this trend, perhaps the further enhancement of new products and education by vendors may make a difference. Complete the SWPP Summer Survey on page 3 or online at and you will receive the complete results of the survey.

15 I n d u s t r y N e w s Concerto Software and Aspect Communications to Combine Westford, MA and San Jose, CA Concerto Software Inc. and Aspect Communications Corporation (NAS- DAQ: ASPT) announced a definitive agreement to combine the companies. The transaction will form the largest company solely focused on contact center products and services. The merger combines Concerto s leadership in predictive dialing and unified contact center systems with Aspect s leadership in contact center workforce management applications and performance analytics. In addition, led by the companies innovative traditional voice and voiceover-internet-protocol (VoIP) automatic call distributors (ACD), the new company will be able to offer a rich product portfolio that includes multi-channel routing, self-service interactive voice response (IVR), Internet contact and virtual contact center capabilities, reporting, monitoring and recording. This portfolio will present customers with both a unified and tightly integrated approach for executing on their customer service, sales, and collections strategies. Frost & Sullivan Honors IEX with Growth Strategy Leadership Award Richardson, TX IEX Corporation, a Tekelec company (Nasdaq: TKLC), announced it has earned the 2005 Growth Leadership Award presented by Frost & Sullivan in the World Agent Performance Optimization Markets report. IEX also received the number one ranking in market share of workforce management software licenses for the fourth year in a row. A well planned and executed growth strategy built upon a strong and deep product portfolio, coupled with an insight into industry and market trends and drivers have been some of the factors responsible for IEX s excellent growth year in Frost & Sullivan lauds IEX s successful growth in a maturing and consolidating marketplace, the report stated. Frost & Sullivan mentions that in a market experiencing saturation in large contact centers within North America, IEX outperformed the competition by registering the best year in the company s history. Increasing revenue by 16 percent over 2003, IEX reinforced its stronghold in the workforce management market. GMT Wins Contract with Amegy Bank to Provide Workforce Management Software Atlanta, GA GMT announced that Amegy Bank N.A., the largest independent bank in Houston and one of the fastest growing banks in Texas, has selected the GMT PlanetTM workforce management software solution to maximize efficiency throughout its Retail bank. After initially searching for separate vendors to provide workforce management for their more than 75 banking centers and contact center, Amegy Bank found GMT s approach provided a comprehensive solution to help meet its needs. After nearly a year of due diligence, our selection team agreed that GMT Planet provided the best solutions for our needs, said Kay Yeager, Senior Vice President and Retail Management Support Manager for Amegy Bank. Enhancing the customer experience is very important to us. GMT Planet will help us direct the right resources to the appropriate areas by effectively forecasting transactions by type, volume, and location. Additionally, it will provide an easy to use reporting and analysis tool for retail management and banking center associates. InVision Software Launches US Subsidiary 15 Ratingen, Germany InVision Software, Europe s leading provider of workforce management software, has announced it has launched a US subsidiary in Chicago, IL. The new headquarters of InVision Software, Inc. will service both the US and Canadian territories, with further bases planned for the West and South to be nearer to those specific markets. Peter Bollenbeck, CEO of the InVision Software Group, says: We are confident that the North American market, across all industries, holds great potential for our company. Obviously, it is a very large territory, and it is still the place where global technology trends are set. With the strong foothold we have in Europe, demonstrated by currently 480 installations, InVision has what it takes to meet the North American challenge. Dickinson Financial Corporation Selects Advanced Workforce Optimization Solution From Witness Systems Atlanta, GA Witness Systems (NAS- DAQ: WITS), a leading global provider of workforce optimization software and services, announced that Dickinson Financial Corporation (DFC), the holding company for Bank Midwest and its affiliated banks, has invested in its advanced workforce optimization solution. The deployment will take place in the Missouri and Kansas contact center operations for one of the institution s subsidiaries. The financial institution, whose contact centers handle standard account transactions, real estate and consumer loans, and global banking, selected Witness Systems based on its pre-packaged workforce optimization software and services solution. After thorough evaluation of several competing technologies, Dickinson Financial Corporation recognized the value in working with a single vendor solution, which will offer the company a comprehensive workforce optimization solution that brings together four functional areas: quality monitoring/contact center recording, workforce management, performance management, and e-learning.

16 Join the Society of Workforce Planning Professionals Become a part of an organization designed specifically to facilitate professional development of workforce planning professionals. SWPP provides its membership with a variety of benefits, including this quarterly newsletter, regional networking meetings, online forums, an annual conference, and more. Membership in SWPP is available to anyone in the workforce planning or related profession. There are three distinct levels of membership in SWPP: individual/associate membership, site membership, and corporate membership. One of these memberships is right for you! Membership costs vary with the type of membership. Prices for each membership are as follows: Individual/Associate Membership $295 USD Site Membership (up to 3 members) $595 USD Corporate Membership (unlimited number of members) $4995 USD Membership applications are available on the web at Still have more questions? Call us at or us at info@swpp.org. We d love to hear from you! SWPP Summer 2005 A quarterly publication of the Society of Workforce Planning Professionals, 6508 Grayson Court Nashville, TN Editor: Vicki Herrell vicki.herrell@swpp.org 2002 Society of Workforce Planning Professionals Society of Workforce Planning Professionals 6508 Grayson Court Nashville, TN Visit us on the web at