Call Centres
...as Gordon Mac Pherson pointed out in The Great Call Center Brain Teaser, is that these are high level output measurements, and focusing on them won’t inherently improve them. To make improvements, we have to improve the factors that cause them to be where they are. Second, as with any measure, we must ensure that they are as accurate, complete and unbiased as possible. Finally, they should be interpreted in light of how they relate to each other; by themselves, any can lead to erroneous conclusions, but together they paint a fairly complete, high-level picture of call center performance. Third, leading call centers have cultivated a systematic, collaborative approach to call center planning. This process generally includes seven major step: 1. Choose an appropriate service level objective 2. Collect necessary planning data 3. Forecast the calling load 4. Calculate the on-phone staff requirements 5. Calculate trunk requirements 6. Factor in roster staff factor or “shrink factor” (which reflects breaks, absenteeism, etc) 7. Organize schedules They have also created a flow chart that illustrates the process step by step and shows the logical sequence of events. The flow chart high lights any “disconnects” in the planning process, such as the marketing department running campaigns that the call center doesn’t know about ahead of time. Systematic planning contributes to effective communication in several ways. It creates a body of information that wouldn’t otherwise be available (“here’s our call load pattern and, therefore, why the schedules are structured as they are”). It also forces people to look into the future and see their work in the context of a larger framework. Perhaps most important, formal planning requires communication about values, on issues such as resource allocations, budgeting and workload priorities. In sum, as one call center manager put it, “formal planning goes far beyond getting the right number of butts in seats. “If forces the kind of communication that an active call center desperately requires. Finally, leaders of high- performance call centers recognize an interesting paradox: too much communicating inhibits effective communication. Farson maintains that “there seems always to be an optimal level of communication beyond which further or expanded communication becomes dysfunctional. Communication has it’s limits.” Too many meetings, memos, conferences, electronic mail messages and on-the-fly discussions may be symptoms of weaknesses in the process. With better tools, more focused training, and appropriate levels of empowerment, the need for excessive communicating can be avoided. I once heard someone draw an analogy to a crew on a sailboat. When one of the lines breaks, nobody waits for anyone else to act, and nobody needs to give orders or instructions. The members of the crew are acting in harmony and know what to do to address the problem. Conclusion Effective communication is inseparable from effective leadership. As Warren Bennis puts it, “Leaders are only as powerful as the ideas they can communicate”. Effective communication results in a shared vision. And, when people are aligned behind a set of compelling values, enthusiasm and commitment— that perceptible energy— tends to follow. 7. Call Centre Typical System Requirement The following is the typical system architecture of a modern call center. SYSTEM REQUIREMENTS End-User/Customer Environment · Operating System: · Microsoft Windows 95, Windows 98 or Windows NT 4.05, supporting voice calling, text chat, E-mail, fax, and callback. · Java- and JavaScript-enabled7 Web browser such as Internet Explorer 3.0 or Netscape Navigator 3.0, or later · Internet access: – Minimum 14.4 kbps for text chat, E-mail, and fax – Minimum 28.8 kbps for voice calls · For voice calls: – To use the Internet for voice calling, the PC must be multimedia-equipped, with an H.323-compliant telephony application such as Microsoft NetMeeting 2.08, or later, and a high-speed connection to the Internet (minimum 28.8 kbps). The H.323- compliant telephony applications currently operate with Windows 95 or Windows NT.5 – Sound card, headset or speakers, and microphone The (Internet Telephony Gateway) ITG generally downloads a call control applet, written in Java, to the customer’s PC to start the Internet telephony application. (An applet is a "mini" application program that is created in the Java language and run on a PC using a Javaenabled browser.) The call control applet provides the customer with call status messages as well as the interface through which the customer drops the call, conducts text chat, or types in data to collaborate with a call center agent. Customer uses the web page interface to request the type of call he/she wants, an internet voice call, a text chat or even a call back or an ordinary telephone using PSTN. Call Center Environment · Internet Telephony Gateway (ITG) (for Voice over IP) · Enterprise Communications Server (ECS) which also supports phantom calling for e-mail/fax queuing and routing. · Automatic Call Distributor ( ACD ) to queue and route Internet calls, E-mail and fax. · Internet Call Manager (ICM) Software · Expert Agent Selection (EAS) routing software manages routing of the call to the best available agent, at any location, with the proper skill level, based on EAS. · Call Management System (CMS) reveals information pertinent to market analysis and sales and service strategies, as well as to the effective management of the call center. Internet call statistics such as average talk time, queue time, the numbers of calls, and so on, are collected by the CMS and reported the same way that other call statistics are gathered. CMS also gathers data on calls that could not be delivered to the ACD due to insufficient ITG resources or because no agents were staffed/logged into the application. · High speed LAN connectivity and Adjunct Switch Application Interface (ASAI ) provides the links for the ECS, computer telephony server, ITG, mail server and Agent PC's. · ISDN-PRI software-the ITG launches the call across the PRI facility to the ECS, where it is queued according to its Vector Directory Number. · Message call application software · POP-3 compliant mail server and software (for E-mail) for example Intuity Audix ( R) with Internet messaging. · Computer Telephony software · Real-time communications software · Messaging software (for E-mail) Agent Desktop · Operating System: – Windows 95, Windows 98, or Windows NT 4.0, Sun OS, Solaris, MacOS 7.1 or later, or OS/26 · Internet Explorer 4.x or Netscape Communicator 4.06, or later · ECS-compatible voice terminal The above typical system architecture of a call center will provide businesses with the opportunity to reach customers with personalized, responsive sales and services. 8. Call Center Planning I) Introduction In planning an inbound sales call center, various questions to be addressed are: 1. What's the call center for? Will it be used for post-sales service and support? If so, will the same people be making both inbound and outbound calls? 2. The expected number of calls; their duration; the hours and days of operation. 3. What industry-specific needs does your center have? 4. What kind of organization & management structure will the center have, and what kind of data and reporting will those executives need? 5. What alternate access channels does your center have to be equipped for: email, web, chat, fax, etc? 6. Are there enough people available to staff your center for the next 4 to 5 years, even with a turnover rate exceeding 30%. 7. What budget has been provisioned to implement these now and/or a year from now? 8. To what extent there is a resource commitment in technology? 9. What telecom and data networking resources are going to be needed to create the center? Presumably these already exist in some form as well, and need to be fitted into the new model of the center (that is, customer history data, product information, web sites, incoming phone numbers, etc.). This list is not exhaustive and there are many more questions to be asked depending upon the location, competition, expectation of revenue generation , or expectation of cost reduction etc. Any major call center decision should be taken only after you've thought through all of these very carefully, and many more besides. A systematic, collaborative approach is required for call center planning. This process generally includes seven major steps: 1. Choose an appropriate service level objective Service level takes the form of X percent answer in Y seconds (such as 80 percent of calls answered in 20 seconds), and is a high level measure of how fast callers get through to agents. A well managed call centers take service level seriously, and strive to meet it as consistently as possible. An appropriately selected service level objective would optimise costs. Answering calls less quickly (or a lot more quickly) would actually cost more money, not less money . There is generally no "industry standard" service level. Different industries have different standards. And different kinds of callers and centers within an industry will have varying standards. There is no consensus on how many calls should be answered within any stretch of time, largely because call centers fulfill so many functions within different industries that you rarely end up comparing apples to apples. There is no magic figure that can tell you whether your particular center is performing better or worse than the universe of all call centers. The correct service level for your call center should meet the following criteria: · Meets caller's needs and expectations · Keeps abandonment at an acceptable level · Minimizes expenses · Maximizes revenues · Minimizes agent errors and maintains quality of service. The optimum service level is affected by a myriad of factors, including the value of the call, fully loaded labor costs, trunk costs and caller tolerances. The service level (one of many metrics that determines how well the center is performing relative to its customers) should be set by the company, using its own standards of what's acceptable behavior. If you are in a competitive industry (e.g. shipping, mutual funds, catalogs etc.) and want to be on the high-end scale, 90/20 is fairly common. Engineer for no more than one percent blockage on the trunks. If you hit these targets reasonably well, your abandonment rate will probably be around one or two percent. 2. Collect necessary planning data As mentioned in the introduction section necessary planning data such as the expected number of calls; their duration; the hours and days of operation, industry specific needs, organization and management structure, availability of budget, etc. is to be collected. 3. Forecast the calling load How can we improve the accuracy of our forecast? Incoming call center management is the art of having the right number of people and supporting resources in place at the right times to handle an accurately forecasted workload at service level, and with quality. Accurately predicting the workload presents one of the most important, and often most challenging steps in this effort. It is a colloborative process where cooperation of all the departments like marketing, sales, manufacturing is required in making known significant developments in their respective departments to other departments. In the absence of a such a colloborative process highly trained person ( (or group) who does the forecasting, equipped with the latest in forecasting software, and armed with every conceivable ACD and database report will still be unable to produce good forecasts Following eleven ways improve the predictability of the workload thus leading to an accurate forecast.: a) Use ACD modes consistently. Each rep has an impact on the components of handling time (talk time and after call work) and, therefore, on the data that will be used in forecasting and planning for future callloads. When the queue is building, it can be tempting to postpone some after call work (wrap-up) that should be done at the time of the call. It is important to define ahead of time which types of work should follow calls and which types of work can wait. This will help in keeping error levels low. b) Avoid callbacks. Many call centers have discovered the hard way that giving callers the option to leave a message when the queue gets backed up often backfires. For example, you may call back only to get perpetual busies, ring-no-answers, voice mail or somebody else in the person's work area ("sorry, she stepped away for a moment"). And in the meantime, the caller may call the call center again. It is always better to handle the inbound calls when they arrive. c) Emphasize quality. Supervisors and agents can feel that the pressure of a backed-up queue forces them to make tough tradeoffs between seemingly competing objectives, such as service level and quality. However, although service level and quality seem to be at odds in the short term, poor quality will negatively impact service level over time by contributing to repeat calls and other forms of waste and rework. This will contribute to workload volatility and inconsistencies. The emphasis should be on handling each call correctly, regardless of how backed up the queue is. d)Better Coordination of various depatments. The forecast can never be accurate if the call center planning group do not know well in advance when marketing is running the next campaign, when manufacturing is releasing the new products and when finance is redesigning the terms and conditions. e) Anticipate the factors affecting caller tolerance. The seven factors of caller tolerance include motivation, availability of substitutes, competition's service level, level of expectations, time available, who's paying for the call and human behavior. Analysing these factors go a long ways towards anticipating caller behavior. f) Track absenteeism. It is important to anticipate absenteeism in advance and it is reasonably predictable. For example, in work groups with typical Monday through Saturday schedules, unscheduled absenteeism tends to be higher on Monday and Saturday than the other days of the week. g) Make forecasting a collaborative process. Involve supervisors and lead agents in the forecasting process, on a rotating basis. This yield two positive results: 1) they will better understand the pulse of the call-load and what's behind the schedules (and will often adhere to them better as a result), and 2) because they are continually dealing with callers, they have their "ear to the ground" and can help anticipate caller reactions to changes and developments in the marketplace and the organization's services. h) Track and manage non-phone activities Non-phone activities such as research and correspondence are to be tracked as these have a strong correlation to other forecasts, such as the inbound call-load, units of sales or number of customers (and they are usually a lot less time-sensitive than incoming calls). . Investigate the tracking capabilities in your ACD, forecasting/staffing software and computer database. As a last resort, track these activities manually, as they occur. i)Accomplish as much as possible during talk time. When tasks related to inbound calls can be completed with the caller still on the line, errors are usually reduced. j) Minimize transferred and escalated calls. Utilize quality improvement tools, such as flow charts and cause and effect diagrams to address root causes to take care of common problems such as insufficient training, insufficient authority, incomplete or missing database information and poor call routing design (e.g. calls often end up in the wrong place to begin with). These quality improvement steps will reduce the excessive number of transferred and escalated calls k) Callers be made aware of other service alternatives. The inbound call-load tends to be less erratic when callers are aware of other service alternatives (e.g. services via faxback, voice response units or the World Wide Web). Focused advertisements, newsletter articles, and customer support sections in user manuals are all examples of ways to better educate callers on the service alternatives available. 4. Calculate the on-phone staff requirements This graph, based on the Erlang C formula, illustrates the relationship between staff and service level in an incoming call center. It shows that 30 reps at the given call load will provide a service level of just over 23 percent in 20 seconds. These numbers are representative only as at this low level of service , many of the calls may get cleared via busies and abandonments, so Erlang C may exaggerate how bad things will be. It depicts that at this level service will be poor. With 31 reps, things improve dramatically. Service level jumps to 44.5 percent, a quantum improvement. Adding one more person yields another big improvement. In fact, adding only four or five people takes service level from abysmally poor service to something respectable. That, of course, means a commensurate drop in average speed of answer (ASA) and trunk load. Keep adding reps, though, and the marginal improvements in service level progressively decline. This is an example of what economists have termed the "law of diminishing returns." This understanding of the relationship between varying levels of resources and service level helps in determining the service level to be aimed during the budgeting process. Those who want to be the "best of the best" in terms of service level find that it increases substantially the expenses towards the staffing budget. Also if you have the right number of people handling calls to begin with, but just a few of them unplug or go "unavailable" at an inopportune moment, things begin to back up. Think of what a stalled car blocking just one lane can quickly do to a busy expressway. 5. Calculate trunk requirements Trunk is an Exchange line or telephone circuit linking two switching systems.The two factors to be taken into consideration are ratio of trunks to extensions which in the case of PABX is usually between 1:5 and 1:10 and Queue calculations. Both Erlang-B and Extended Erlang-B are supported for trunk requirement calculations. 6. Plan for roster staff to take care of breaks, absenteeism, etc 7. Organize schedules. Organizing schedule means as to (a) when and for how much time agents are able to take calls during their shifts including the time spent handling calls and the time spent waiting for calls to arrive (also called Availability), and b) when they are made available to take calls (also called Compliance or Adherence). Adherence to schedule is the term that refers to how well agents are able to adhere to their schedules. Conclusion: It always seems that no matter how well you plan, there are always either more resources than demand or more demand than resources. The truth is, real-time management needs to compliment good planning. Your on the spot decisions from time to time are what enable you to maintain your service level. Managing your service level in real-time has to consider following points. · Establishing a good foundation ahead of time so that you are not creating many of the crises to which you are responding. · Providing real-time information to agents and supervisors, and training them on how to interpret the information. · Planning your escalation procedure ahead of time and defining the thresholds that will determine when alternatives are deployed. · Continually reviewing your planning process. Real-time management will never be an effective substitute for accurate resource planning. Thus effective real-time management takes planning, coordination and practice. 9. CALL CENTER- WORKFORCE MANAGEMENT Telephone traffic along with other forms of traffic, such as automobile traffic, share similar characteristics. Traffic may be very busy and have to slow down and wait, or it may be very light with little slowing or blockage. Facilities such as roads, telephone lines, toll booths, service agents, and bank tellers may be either under or over utilized causing costly idle time or poor service to customers. Call center managers face a daily, continual challenge in finding the right number of agents to use at the right time to handle calls that arrive in a random manner. A. The Problem B. How Calls arrive in a Call Center C. The Two Solutions C1. Formula Solution C2. Simulation Solution D. What Method is best for your Call Center E. Workforce Management Tools A The Problem The fundamental challenge for the inbound call center manager is to correctly balance the number of agents and trunks to the varying volumes of calls throughout the day in order to: 1. Keep all the agents busy, and 2. Keep the time callers have to wait to an acceptable minimum. Calls, arrive in the system to be served by agents or operators, but may have to be in queue for them. There is a long past history of mathematical study which has gone into the problem of "arrivals" that need service by a "server". B How Calls Arrive in a Call Center It is fact that call arrivals in the real world are not distributed one right after the other. The actual arrival of calls is distributed randomly: some will come in at the same time, some will come in when another is being served, and during some periods of the hour no calls may arrive at all. Also, the length of the calls follows a random distribution. These random distributions are determined by the laws of probability. How then can the actual arrival rate of calls in your call center be predicted? While we cannot predict the exact time that each call will arrive, we can predict the probability of when the next call will arrive. This probability follows a well known distribution called the Poisson distribution. It looks something like the following bar graph: Probability of Arrival Minutes to next call arrival average minutes to next call arrival It's a bell-shaped distribution skewed out to the right. This means in practice considering the mathematics behind the Poisson distribution that call arrivals will always tend to be clumped together and will not arrive in an even manner. The graph of observed data for % Agent Utilization has a gentle slope, but graphs for Average Speed to Answer and Average Time in Queue have very steep slopes. Thus, the relationship between how busy your agents are and the service your callers get is not a simple linear relationship. There are break off points where the addition or subtraction of one or two agents will result in dramatic differences in customer service level. The lengths of calls are not uniform either. Call lengths nearly always closely follow what is called the exponential distribution: This distribution says that the most probable call lengths are those that are less than the average call length, but that there are going to be some that are a whole lot longer. This phenomenon also contributes to a "clumping effect": the length of calls in progress at any given time may all be rather short or all be rather long, but will very seldom be close to the average length of all the calls experienced in a day. This is further complicated by the fact that the average time that callers spend in queue is also exponentially distributed around the average queue time primarily because a large percentage of callers hang up very quickly when they get the "please hold" message. C The Two Solutions Using the assumptions of the above distributions for call arrivals and call length, and the customer/server/queuing nature of an inbound call center, there are two mathematical methods that can be used to optimize a call center’s performance: 1. A formula method that calculates a statistical equation to predict queuing times, agent workloads, and optimal agent levels, or 2. A simulation method that imitates the flow of calls into a call center and their handling by agents, with queuing, blocking, and overflow. C1. The "Formula" Solution The most common method used for call center staffing is formula using the famous Erlang equations (Erlang C and Erlang B) for staffing and trunking. They compute the probability of delay in queue for calls when using a given agent level with a given amount of inbound call traffic. The software in the various call center staffing and scheduling products that calculate the Erlang C formula calculate it for several agent levels and then pick out and tell the user the one agent level that gives a delay probability closest to that of their desired service level. These products will do this for each hour, half hour, or quarter hour time period throughout a workday. So, in the formula method, the user inputs number of calls, talk times, and wrap-up times for each period of time in the workday, along with a desired service level (e.g., answer 75% of calls within 20 seconds). The software calculates the Erlang equation, and pops out optimum agent levels for all time periods in the day. It can also be used to observe the effects on queuing and agent workload when using different agent levels in any one time period. In fact, majority of the staffing and scheduling software for call centers use the Erlang formulas or some minor modification of them. The Erlang analytical method has done a good job over the years in helping call center managers determine what agent levels they should use. However, it does not account for some of the unique dynamics of inbound call centers using an ACD, and it tends to overstaff many call centers to some degree. The most talked about shortcoming of the standard Erlang C formula is that it assumes calls will queue up infinitely and wait forever to be answered by an agent. For centers with limited inbound line capacity and callers with little queuing tolerance, the Erlang method may not be accurate enough. Also, what about caller abandonment (hanging up after being queued), or callers retrying after abandoning or getting a busy, or the ACD overflowing calls to another agent gro, or ACD ring delay feature (the number of rings the caller can be given before the ACD picks up the line)? Some companies and mathematicians have derived "special versions" of the Erlang method that claim to at least partially account for some of these issues especially the infinite queuing problem. However, no tweaking of the equation will fully take into account all of the issues mentioned above. Some of the variants out there to the Erlang C equation are highly suspect, having no published analysis of their mathematics, and may actually give less reliable staffing predictions. If a call center manager finds that in practice Erlang C is not allowing them to maintain their desired service levels to the degree they want, or that it calls for many more agents than are actually needed, then they should probably just skip all the Erlang variants and analytical equation methods and try the second method: simulation software. C2 The Simulation Solution First we may discuss what is computer simulation? Most people have heard of weather and climate forecasts made by computer simulations. Industrial and natural processes may often be simulated by a computer program. The program imitates the flow of people, material, or events and is used to experiment and see what the effects that alternative configurations of the people, material or events would have on a real-world situation. In the past few years simulation programs have become an increasingly popular method for call center staffing and performance prediction. A call center or ACD simulator compresses a day’s worth of incoming call traffic into a short period of time by representing each second of a day with a few microseconds of computer time. Calls are made to randomly arrive in a "virtual" call center created in the computer’s memory and are answered by agents, queued, met with busy signals, overflowed, or abandoned depending on the parameters, agent/trunk combinations, and call volumes defined for the call center. The simulator doesn’t just calculate equations in the way that a formula does, rather, it actually acts as an experimental call center that reacts to the random arrival of inbound calls with the agent and inbound line levels, inbound call volumes, caller abandonment and retry behavior, and ACD overflow the call center manager wants to test and try out. Simulation can then accurately predict the levels and effects on service level of call blockage, abandonment, caller retries and overflow when these are entered as parameters to the simulation. Also, a simulation program can make the random call arrivals and call lengths follow the Poisson and exponential distributions that are seen in the real world. At the end of the simulation, you will see how many calls were handled, abandoned, given a busy, queued, and overflowed along with queue lengths, average speed to answer, and the actual service levels that would be given to your callers. More simulations with different combinations of agents and trunks are usually performed until the manager is satisfied with the service levels, agent work load, and call center finances. The following statement highlights the the difference between the Erlang formula method and the simulation method. With the formula method, you enter your call volumes and what you want to happen in the form of the desired service level you want to maintain, and you are told what number of agents and lines you need. With the simulation method you enter the number of agents and lines you want to use along with your call volumes, and you see what will happen in the form of service levels actually given. One drawback of simulation is that it can take longer to determine your optimal agent levels. The Erlang formula’s output is an optimal agent level: you only need to run it once. With simulation you have to test and try out different agent levels and see what happens. Are the resulting queue levels acceptable? Are the agents busy enough? Is the service level sufficient? No? … then adjust the agent and line levels and run another simulation, and another, and maybe another. This extra work pays back in what will be more accurate and practical predictions of optimal agent levels and call center performance. Also you can see values for calls queued and overflowed, the number of callers who abandon or get busies, and the calls that retry later. D What Method Is Best For Your Call Center? If you require staffing predictions to do simple scheduling and just need reasonably accurate levels to on which to base your agent schedules, then working with Erlang C or an Erlang variant will probably fit your needs. Practically all of the existing workforce management systems use an Erlang formula for calculating the required number of agents, and these requirements are then fed into a scheduling algorithm and agent database to create work schedules. Erlang C also works well with most of the home grown spreadsheet based scheduling systems that many call centers have developed independently. Problems occur if your call center is more complex. If you have multiple agent groups, skills based routing of calls, overflow of calls from one group to another, or auto attendant and IVR systems that handle some of the calls, then simulation becomes the better option since these dynamics cannot be represented in a mathematical formula. However, at present, there are hardly any simulation software products that handle these and are combined with a scheduling module in the same package Most call center managers use Erlang and simulation staffing methods for call center performance analysis in addition to agent scheduling. They want to see what they’re predicted queue levels and agent work load will be at varying agent levels. Analysis of call center performance is especially helpful in taking a close look at your call center’s busiest hours, least busy hours, and other problem periods. For performance analysis, you don’t want an Erlang formula tool that just reports a required numbers of agents based on a given service level goal. You want a tool that will also tell you what your average speed to answer, percent queued, average queue length, agent occupancy will be. These and other performance measures are important to study in order to determine what your service level goal should be in the first place. Simulation software will also give these measures for call center analysis and is more accurate. Also, as already stated, simulation can take into account caller abandonment, retry behavior, ACD ring delay, and overflow. On the minus side, call center simulators can be very expen...