|
iOps An insight into BPO Operations |
||||||||||||
|
January 2005 |
|
Issue:2 |
||||||||||
|
|
Operations Intelligence If you can't measure it, you can't improve it. While this statement is powerful, somehow, its author missed out the urgency factor that characterizes real time businesses such as call centers. The statement could be aptly modified as "If you can't measure it Now, you can't improve it Now." Real Time and Historical Data Call center operations run on data, both real time and as historical. While historical data provides information about trends and clues on problem areas, real time data provides information about the state of operations at the current time. If call centers were airplanes, then historical data would be used to build better airplanes and fly them more efficiently in the long run, while real time data would be used to fly the airplane as best as possible right now in the current operating environment. Need for Real Time Data Most operations managers need to balance both types of data to make decisions. However a key point to note is - the higher the availability of real time data, the more decentralized operational decisions can be. At the present time, lack of such data creates the need for senior managers' experience and skill to make decisions. However, if this data were made available to mid level managers along with requisite training, it could mean, for example, that team leaders would be able to make those decisions for which they had to rely on their managers till now. Therefore, this will mean that team leaders can take effective decisions while freeing up time for their managers to plan and manage operations more efficiently (build better planes). Scenario Today The key to providing real time data is, of course, the availability and implementation of relevant technologies. Bits and pieces have already been implemented today. However more needs to be done. Automated Systems: Call centers in India have already implemented (or use) the first level of technologies from which real time operations data can be derived. Examples include Automated Call Distributors (ACDs) and Dialers. Data from these systems are readily available to operations managers, and are sometimes available to team leaders. However, training on the interpretation of this data for real time action has not yet taken priority. Also, most data is in a "raw" format. It is not packaged into a readily consumable format that would enable managers to make quick instant decisions. Therefore, even through real time data is available, managers need to spend precious time on a spreadsheet reformatting the data and calculating composite metrics. Manual Processes: Data from automated systems provides only one part of the puzzle. The second part deals with data from the numerous manual processes that exist in call centers (e.g. call quality monitoring, calibrations, skip level meetings, agent reviews, compliance processes – to COPC etc., training, daily client status calls etc.). For example, consider a common process like quality monitoring. Data about this process is seldom available to managers and team leader on a timely basis. When it is made available, only the raw data is provided. Again, managers and team leaders have to spend precious time analyzing and making conclusions. Therefore, good real time performance management occurs more as a matter of chance rather than as a matter of skill. The rest of this article focuses on a system that deals more with the process of automating these manual processes and making information from them available to managers in a meaningful way. Operations Intelligence Systems A key requirement in the industry is to make operations predictable. Operational requirements need to be met in every half hour interval. With such real time requirements, the industry needs a system that can provide operations data, in an intelligent manner at various levels of aggregation in the fastest possible time. Enter the Operations Intelligence Systems. This system seeks to provide managers with a handle to manage processes in real time by providing them with data on three key aspects. Process Adherence: Are processes occurring when they are supposed to occur? For instance, are the quality monitoring schedules being adhered to? Are call calibrations being done on time? Are performance reviews being conducted when they are supposed to done? Are team quality reviews being conducted on a regular basis? Was the training program that was scheduled for Tuesday actually conducted to completion? Every process that has been set up needs to take place in order for results to be produced. The above statements are just some examples of processes that often experience slippage. The information of the slippage does not reach managers on time. Therefore, an operations intelligence system needs to provide a method for process owners to update the occurrence of a scheduled activity. It also needs to provide a method for managers to receive information about the schedule adherence of processes. This can be done through reports and alerts. Process Quality: The mere occurrence of a process activity (process adherence) is not sufficient. The quality of the activity also needs to be assured. There are multiple methods to determine the quality of a process. One key method includes determining the "scores" at the end of the activity. For instances, processes like call quality monitoring, training, calibrations etc. always have scores reported at the end of the process. The time lag between the completion of the process itself and the reporting of the score needs to be minimum. Also, if the score is not up to the mark (a preset limit), managers need to know immediately. Again, this can be done through alerts for critical processes or through reports. Process Linkage: A good operations intelligence linkage system will also capture linkages between processes. A great example is the linkage between call calibrations and call monitoring. Call calibrations are done in order to assure the quality of the call monitoring process itself. Therefore, if call calibrations are not done on time, or if call calibration scores are not within the prescribed limits, there will be a downstream effect on call monitoring scores. Hence, the system should, at any point in time, display the effect adherence or quality of upstream processes to the downstream processes. Here are a couple of other such examples: Another example of such a linkage is training for a new batch of agents and performance on the call (represented by quality scores and other metrics such as AHT, SPH etc.). An Operations Intelligence System will provide the Power of Now to managers. The moment a process is set up within the system, it creates an urgency among the process owners to adhere to their schedules and to ensure that processes are conducted with the requisite quality. It takes operations a level higher by making it more process dependent rather than individual dependent. Here are some of the key benefits of this system 1. The system will remind process owners to conduct the activity on time. The system can print or display the daily activities of the process owner that must be completed. Therefore, Team leaders will never miss review sessions with their teams because the system compels them to have the review session. Training programs will never be delayed or postponed unnecessarily because the system will compel trainers to conduct sessions on time. Any schedule slippage or quality deviations will give rise to visual indicators that turn from green to red. 2. Managers can get alerted if the activity is not conducted on time. Instead of reactively discovering this fact, the system will make the manager proactive by providing her with data on process adherence. Managers only need to focus on resolving the problems that have prevented the process from taking place or that have caused quality problems. 3. After a process or an activity has been completed, the system will alert the manager if the quality of the process is not within the prescribed norms. Again, the manager can now proactively fix the problem. 4. The system can roll up process results at various managerial levels through the use of dashboards and alerts. For instance, the Senior Manager can be alerted that there is a possibility of process problems within the next month because 30% of the call calibrations have been missed last week. They can also be alerted that all the new agents in the training program will not hit the floor next month because 30% of them have received less than acceptable scores in an intermediate test. This will give rise to a staffing shortage and will eventually hit service levels. 5. Managers can ensure schedule adherence for temporary processes that have been created to tide over a problem. For example, a data gathering effort may be in progress, to capture all customer complaints and discuss them with the client on a call every week. As soon as this process is defined in the system, managers can rest assured that the process owner will be prompted by the system to complete the activity as and when required. Managers will also get data and alerts on a regular basis. Requirement of an Operations Intelligence System: In order for this system to be effective, certain key requirements need to be met. 1. The system must be highly flexible i.e. one must be able to define multiple types of processes within it. For example, some processes will occur in a time based manner while others will occur on an event-based manner. Another example is that some processes may generate metrics while others will generate textual data (e.g. results of the client call in point no. 5 above). 2. The system must have an easy-to-use user interface where managers or process owners can define processes with considerable ease. This will include defining reports and creating an "alerts" system. 3. The system must have reporting capabilities. Process owners should be able to create reports easily and quickly. This includes features like quick analytical tools and emailing capability. 4. The system must have an alerting mechanism that can be set up by recipients of the information. For instance, after the process owner has created a process in the system, the manager must be able to set up an alerting system that can send alerts if some criterion is not met. 5. The system should be able to display information in a dashboard format to managers at various levels. An Operations Intelligence System can put the Power of Now in the hands of call center managers and team leaders. Developing this system will require a fair amount of introspection into operations in terms of identifying processes. Who will develop this type of a system, when and how, remains to be seen. OnTrac Internal Staff |
|
||||||||||
|
To unsubscribe send a blank email to leave-iops-newsletter@iamontrac.com iOps is a newsletter created for Operations Professionals in the BPO and Call Center Industries in India. This newsletter is published by OnTrac, a unique provider of Operations and Management training for Call Centers and BPO companies. We welcome contributions to this newsletter. Please contact us at the phone number or email address given. Our Contact Information: Address:
OnTrac Phone:
+91-80-51261274 |
||||||||||||