Project was aimed to identify primary cost drivers in IT services by focusing on people, process and services & to recommend areas of focus for optimal cost reduction through efficiency improvements, thereby resulting in quantified expected benefits. Also to baseline the process by providing list of identified process improvements for contract negotiations and vendor management.
The project was guided by the following set of broadly stated research objectives:
- Preliminary volume analysis to understand the major call & ticket volume contributors and to have a high level understanding of workflows associated with the present method of operations
- Detailed process mapping to identify the demand drivers and existing workflows aimed to enable the client to visually characterize their key processes
- AIdentification of key factors driving call volume by using statistical tools
- Action planning - communication of the proposed solution
- Standardization of process documentation for future references
Let's study the key phases to understand how we arrived at the solution.
Preliminary Volume Analysis
op 3 sites contributing to 80% of the monthly ticket volume were identified by analyzing the historical data for the past 3 months, using statistical tools like Pareto and ANOVA. Each handled ticket was assigned a Cause Code, Reason & Reason Detail. After an analysis of the top two sites and at the enterprise level, it was found that over 70% of the ticket volume is contributed by 3 cause codes further divided into 8 reasons and 20 reason details. Service levels on ticket handle time were driven by the Severity (1-4, 1 being the Highest & 4 being the lowest) and it was found during the preliminary volume analysis that around 85% of the tickets were marked as Severity 4. The following figure shows the levels of analysis:
Detailed Process Mapping
The process workflows for each of the 20 reason details have been mapped in detail, separately for each vendor based on the following logical distribution:
With the help of the detailed process maps, the Turnaround Time (TAT) was analyzed to understand the variation between the two vendors and it was found that one of the vendors was taking lesser time to handle a ticket at Tier 2 & 3.
Key Factor Identification
The TAT was further drilled down to identify the statistically significant factors for each vendor. These factors were then discussed in a brainstorming session with the vendors to find out the top reasons with the help of statistical tools like Cause & Effect Analysis and Control-Impact matrix. The results are as below:
Current Model Restructure: Vendor 1 had a fairly smaller time spend (or touch time) to process the ticket as compared to that of Vendor 2. With ticket cost being a function of time spent, it was analyzed that Vendor 1 had staffed more resources than required and was paid a very high ticket cost. Moreover, Vendor 2 was not consistently measuring their time spent which led to an assumption of a higher (than usual) figure for time spent in the TAT analysis.
Even though process model restructure can potentially increase the TAT by a few hours however the gap between current and target TAT allowed the staff to be reduced considerably as shown in the following control charts (X-Bar R)
Process Optimization through TAT Variation Reduction & Process Standardization: The data dependencies on user community inflate the overall TAT by 5-8 business hours. This also indirectly impacts the touch time as the technician takes time to update the records or communicate with the user to explain the need of required data. Recommendations were provided on the following key factors deduced during the analysis, which were responsible for high variation in TAT:
- User unavailable to confirm the ticket resolution
- Lack of user training on ticket logging
- Inadequate / incorrect details supplied by the user in the logged ticket
- Approver unavailability for approval request
- Lack of documentation for older applications supported
- Additional time spent in cases of new issues (cause codes / reasons / reason details)
- Inability to detail out issues in ticket logging application
- Technician knowledge on new applications / issues
Recommendations: Through the following recommendations the process optimization would result in reduction in Turnaround time variation, and thereby making the process more predictable and improving the productivity:
Our efforts resulted in providing a complete set of optimized process and potential annual savings of $ 6Million ($ 515k per month). It also led to a significant reduction of 40% in their current ticket cost. The process model restructure would potentially allow the staff to be reduced considerably. There was a comprehensive report shared with the stakeholders within a period of 8 weeks and was hugely appreciated for the recommendations on the strategic placement of vendors as their current managed service agreements appeared to have structures those were limiting /unfavourable to the company's interests. Process optimization would result in reduction in variation in TAT business hours thereby making the process more predictable and improving the productivity.