Question

Read the case Laureani, A., Antony, J. (2009). Lean six sigma in a call center: a case study. Emerald Insight found in the Unit 3 activities checklist.

Describe the tools used in each one of the DMAIC phases in this project. Create a table with that information. What tools you think may have been redundant? What other tools you may have used to complement each of the DMAIC phases? Include costs, DPMO, sigma levels, and process capabilities if available. Write your conclusions.

Case study The case study focuses on a large corporation in the service sector operating in the vehicle leasing and renting industry: it sells its services to both the general retail public and other businesses in 145 countries and has 22,000 employees and 8,000 locations. Its European call centre was receiving an average of 10,000 calls a month from customers who had an issue with either the level of service received or the billing/invoicing process. The company puts a lot of emphasis on its level of customer service, an integral part of its mission, vision and values, as a differentiator from the competition; its strategic position was that of a quality service, intended to attract those customers who wanted to pay a bit more for a better service, in contrast with its low-cost competitors. Consequently, quality of customer service was a critical component of the company’s strategy: the number of calls received and the more frequent issues reported by customers were continuously analysed and employees’ variable compensation linked to it. The number of contacts received from customers amounted to about 2 per cent of the total transactions (five million a year). The specific objective of the inbound call centre was to deal as quickly and efficiently as possible with those customers. Problem statement A consistent number of the calls of the 1,200,000 received annually at the inbound call centre couldn’t be solved at the first attempt, leading to customer dissatisfaction and unnecessary repetition of work in the Centre. The objective of the project was to increase the first-call resolution ratio. Lean six sigma in a call centre 761 Define phase A cross-functional project team was created, led by a black belt, with the objective of using the define, measure, analyse, improve, and control (DMAIC) six sigma breakthrough methodology in order to increase the first-call resolution ratio. The team scoped down the project, identifying which specific areas of the call centre and services they were going to focus on; a high-level process map or suppliers, inputs, process, outputs, customers (SIPOC) (Pyzdek, 2003) was created, followed by a more detailed process map (see Figures 1 and 2). The team also investigated the seven kinds of waste accordingly to lean principles. These are: (1) overproduction; (2) inventory; (3) motion; (4) waiting; (5) transportation; (6) over-processing; and (7) defects. Of these seven wastes, four were identified as belonging to the current state of the process: . Motion: unnecessary movement from the call centre operators, who needed to leave their workstation to perform some routine tasks, e.g. sending/receiving a fax, printing a document; . Waiting: sometimes an operator could not answer a customer’s query immediately because they were waiting for an answer from another department or a supervisor; . Over-processing: some of the queries from customer could not be addressed on the phone and needed to be dealt with off line by another operator, so creating unnecessary re-work and reducing customer satisfaction; . Defects: a query was not always completely solved the first time, forcing customers to call back and so creating unnecessary work and lowering customer satisfaction. Figure 1. SIPOC (high level process map) IJPPM 59,8 762 Figure 2. Detailed process map Lean six sigma in a call centre 763 To complete the define phase, the operational definition of first-time call resolution was developed and agreed with the major stakeholders: a call was deemed not resolved if, at the end of it, the case was still open in the customer relationship management system, or was escalated or transferred to another department. Measure phase In this phase the project team established the key metrics for the process and calculated the baseline performances of the process at the start of the project. The defects per million opportunities (DPMO) value of the process was calculated (Pyzdek, 2003), using random sampling techniques (see Table I): establishing this is a key component of the Measure Phase, as it becomes the benchmarking performance measures against which the project will be evaluated. It is important to note at this stage that, whatever the measurement system agreed upon, it must be kept consistent at the end of the project, so the pre- and post-project performances of the process can be compared. Analyse phase The team went into the details of the data, slicing them across different dimensions, and noted, using a Pareto chart (see Figure 3), which two types of queries accounted for about 70 per cent of the unresolved first-time calls. Hypothesis tests showed no relationship between the number of unresolved first-time calls and the call centre operators: in other words, there was no statistically significant difference in the number of unresolved first-time calls between more experienced and less experienced operators. As a result, the focus was put on eliminating the root causes for categories “A” and “B” of the queries. Improve phase While collecting and analysing the data, the project team had the chance to “walk the process”, i.e. observe the process in action: this allowed them to identify two so called “quick win” actions (actions that can be easily implemented without roadblocks or concerns). Two improvement ideas were also generated in a brainstorming session. All those actions were first pre-tested through a pilot group, in a sub-section of the overall call centre: further data were collected from the pilot group to quantify the effects of the improvement actions. The sigma value for the new process, after the improvements were implemented, was recalculated in the pilot group (see Table II). Based on the sample size results on the pilot group, the improvement actions reduced the percentage of unresolved first-time calls from 11.82 per cent to 8.45 per cent. This was deemed satisfactory and the improvement actions were rolled out to the whole call centre. Total number of calls received 91,134 Unresolved queries after first call 10,769 Unresolved as percentage of total calls 11.82 DPMO 118,167 Table I. DPMO value calculations at the start of the project IJPPM 59,8 764 Control phase As part of the control phase, the project team documented the improvement actions implemented in a control plan, which was then handed over to the process owner to ensure sustainability of the process going forward. A p-chart was used to monitor and sustain the gains (see Figure 4). The team also calculated the financial impact of the project. Calculating this in a call centre can be tricky, due to the confusion normally existing between so called “hard savings” and “soft savings”: the first are those financial values identifiable in the profit and loss account and balance sheet of the organisation (e.g. payroll costs, telecommunication costs, hardware, software), whereas the latter are not listed in the balance sheet (e.g. productivity increase, customer satisfaction). The 3 per cent decrease in unresolved queries after first contact resulted in 36,000 fewer calls to the call centre on an annual basis (36,000 ¼ 3 per cent of 1,200,000, as this was the number of calls expected to be handled annually). This freed a certain amount of resources, in terms of both staff and telecommunication equipment, whose value was estimated at about $200,000 per annum. Total number of calls received 27,385 Unresolved queries after first call 2,315 Unresolved as percentage of total calls 8.45 DPMO 84,535 Table II. DPMO value calculations after improvement actions are implemented Figure 3. Pareto chart of unresolved first-time calls by type of query Lean six sigma in a call centre 765 Managerial implications and key lessons learned The main managerial implications from the project were in terms of management’s approach to lean six sigma and communication. The success of this particular project made senior management aware of the opportunities the application of lean six sigma created in the call centre environment. Better utilisation of resources, reduced operational costs and improved customer service are vital factors for the long-term sustainability of a call centre: plans for further lean six sigma projects were prepared. In terms of communication, the most relevant aspect was related to employees’ informal communication: the member of the cross-functional project team, working along with the black belt, were exposed to some of the lean six sigma tools and could see their impact on the understanding of the process. Some of them went proceeded to apply for six sigma green belt training, fostering a culture of continuous improvement in the organisation and getting more staff involved in the lean six sigma program. The key lessons to be learned from this case study are related to the first phase (define) and the last phase (control) of project: . The fact that no stakeholder analysis was performed in the define phase may not have caused major problems in this particular project, where the call centre was physically in one unique location, but given the spread of operations across the world, the importance of conducting a proper stakeholder analysis at the outset of the project must be emphasised. . In the control phase, the completed control plan was handed over to the process owner to ensure sustainability of the process, going forward. However, it may be necessary for the black belt, after a certain time, to revisit the process, to ensure improvement actions are still in place, and the process had not reverted to its pre-project status. Figure 4. P control chart IJPPM 59,8 766 Conclusion Call centres are increasingly important for many businesses and are consistently struggling with the pressure of delivering a better service at a lower cost. This paper tried to address the issue of whether lean six sigma can be useful in a call centre environment: by means of a case study, it was found that lean six sigma can improve the operation of a call centre, through an increase in first-call resolution (that reduces failure created by failing to answer the query in the first place), a reduction in call centre operators’ turnover (leveraging on training and experience) and streamlining the underlying processes by eliminating unnecessary operations. The case study illustrated within this paper has used a certain number of the tools available in the lean six sigma toolbox, while omitting some other important tools (e.g. stakeholder analysis and measurement system analysis): however, despite this, it managed to improve the first-call resolution, and hence the customer service levels, of its operation. Given the large scale of many call centre operations, even a relatively small improvement in the sigma value of the process can dramatically reduce the defect rate, increase customer satisfaction and deliver financial benefits to the bottom-line. By focusing on eliminating waste, identifying the truly value adding activities and using the DMAIC tools for problem solving, it is possible to achieve significant improvements in costs and the levels of customer service provided.

Answer 1

Defining phase

A cross-functional project team headed by a black belt was developed with the goal of using the six sigma breakthrough technique (DMAIC) to identify, calculate, evaluate, optimize, and monitor (DMAIC) to improve the first-call response ratio. The team surveyed the project, determining the different call center areas and facilities they should work on; a high-level process map of sources, inputs, systems, outputs, consumers (SIPOC) was generated (Pyzdek, 2003) accompanied by a more comprehensive process chart.

Four were listed from the seven wastes as contributing to the current state of the process: Motion: excessive movement by call center operators who had to leave their workstation to execute certain routine duties, such as sending/receiving faxes, printing a document; Waiting: often an operator did not respond directly to a customer's question while they were waiting for a response from another department or supervisor; Over-processing: some of the customer's questions could not be answered on the phone, so another provider had to deal with them off-line, thereby causing excessive rework so decreasing the customer satisfaction; Defects: The first time a problem was not always entirely answered, causing consumers to call again and therefore generating excessive stress and reducing customer loyalty.

High-level process map

To conclude the specified process, the first-time call settlement procedural concept was established and negotiated with the major interested parties: a call was considered unresolved if, at the conclusion of the call, the case remained open in the client relationship management program, which was either elevated which moved to another agency.

Measurement phase

During this phase, the project team defined the main process parameters and measured the process's baseline output at the project launch. The method value of the faults per million opportunities (DPMO) was measured (Pyzdek, 2003), using random sampling methods (see Table I): defining this is a core component of the Measure Process since it is the benchmarking success metrics by which the project can be evaluated.

Around this point, it is important to remember that whatever the measurement method agreed upon, it must be kept constant at the end of the project, so that the process's pre- and post- success can be measured.

Analysis phase

The team went through the data specifics, moving them over various scales, and observed using a Pareto map (see Figure 3), which accounted for around 70 percent of unresolved first-time calls by two forms of queries.

Hypothesis experiments found no association between the number of first- unresolved calls and call center operators: in other terms, there was no statistically meaningful gap between more seasoned and less experienced operators in the number of first- unresolved calls. As a result, the emphasis was on eliminating the root causes of the queries 'categories "A" and "B."

Improve phase

The project team had the ability to "walk the cycle" when gathering and reviewing the data, i.e. observing the process in action: this helped them to recognize two so- "easy win" steps (steps that can be quickly carried out without roadblocks or concerns). In a brainstorming session, two ideas for change were also produced.

In a sub- of the overall call center, all these activities were the first pre- in a pilot group: additional results were obtained from the pilot group to measure the impact of the enhancement activities.

The sigma value for the current method was recalculated in the pilot community after the changes were introduced (see Table II).

Depending on the pilot community sample size findings, the optimization interventions lowered the number of unanswered first- requests from 11.82 percent to 8.45 percent. This was found acceptable and the measures for change were carried out to the entire contact center.

Control phase

The project manager reported the progress activities undertaken in a management plan as part of the monitoring phase, which was then turned back to the service owner to ensure the procedure is continued.

To track and maintain the gains a p-chart was used (see Figure 4).

The team has also estimated the project's financial impact. Regardless of the confusion that typically occurs between so-called "hard savings" and "soft savings," calculating that in a contact center can be tricky: the former are all financial metrics visible in the organization's income and expense account and balance sheet (e.g. payroll expenses, telephone expenses, infrastructure, software), while the other is not reported in the balance sheet (e.g. pr).

The 3% drop in unanswered requests following initial contact resulted in an estimated 36,000 fewer calls to the call center (36,000 1⁄4 3% of 1,200,000, as that, was the number of calls projected to be treated each year).

It provided a considerable amount of cash, both in terms of personnel and telephone infrastructure, valued at about $200,000 a year.