Q&A: Getting the Most from Your IT Budget
Where should IT look for cost savings, prioritize projects, and estimate your potential savings?
If your shop is like most IT shops, your IT budget is tight and you're being asked to do more with less. Where can you find savings within your IT organization? For suggestions about where to look and how to approach cost savings, we turned to Rose Rowe, vice president of mainframe solutions at Compuware. She started her career as a business expert in the banking industry, moved into IT and held several positions which led her into becoming a consultant and then the vice president of business applications development and support for a major bank. She draws on this diverse background to address topics such as saving money on the mainframe platform from a unique viewpoint.
Enterprise Strategies: What do you see as the biggest challenge in IT shops today?
Rose Rowe: IT organizations are routinely challenged to remain agile and responsive to the needs of their business partners. However, with the current economic climate, they are now expected to meet those needs while holding or cutting costs. That means making the most efficient use of existing resources: hardware, software, and personnel, with attention to business process improvement, operational efficiency, and risk management.
What areas do you suggest IT examine first when they're looking to save money quickly?
By doing a quick software asset inventory review, it is not unusual to find duplication or overlap in tools. Also, having tools in place that are integrated and use common supporting technology, such as source listing files, reduces the time and resources required to use and support them. Both of these items can result in rapid savings.
IT leadership should also look closely at “business as usual” with a focus on opportunities to increase efficiency, improve quality, and to identify poorly performing or resource intensive-applications. Doing so can often result in a variety of savings opportunities including: reduced costs for outsourcers or contractors, elimination of penalties for missed SLAs, postponement of expensive hardware/software upgrades, minimization of risk and costs associated with exposure of sensitive data, and perhaps even avoiding hiring to replace retiring staff.
What areas do you suggest IT examine if they're looking for the largest savings? Which approach do you recommend?
Avoiding or postponing hardware upgrades (and the resulting associated software upgrade fees) is clearly an opportunity for very substantial savings. Organizations can approach this from several different angles through proactive initiatives to:
- Identify and remediate applications that are consuming excessive resources during execution
- Identify applications that are chronically abending, consuming unnecessary CPU resources in reruns
- Include performance analysis as a part of the routine QA
- Examine production promotion processes to avoid unexpected performance degradation
What are typical savings IT can realistically expect?
Savings come in many forms. Results among our customers vary greatly and hinge on many variables including the size of the IT organization, current standards and processes in place, and management support for improving quality and operational efficiencies. We have found that customers experiencing the greatest savings are those who have a proactive focus on quality, application performance, and resource efficiency, and who have best practices, tools and processes in place.
Recent successful savings initiatives that our customers have shared with us: $400,000 annually savings on SLA penalties, 50 percent reduction in MIPS growth returning three Euros for every one invested in tuning, delayed CPU upgrade for 18 months saving $2 million, 60 percent reduction in troubleshooting time, 63 percent reduction in abends.
How should IT prioritize money-saving projects?
Money-saving projects with the potential to provide substantive savings on an ongoing basis should be given top priority. What may at first glance appear to be an opportunity for quick savings in the short term should be scrutinized; IT should ask: aside from the price tag, what are all the costs associated with implementing this approach, such as conversion cost, lost productivity, and training costs?
How do you estimate potential cost savings? What metrics do you consider?
Any estimates for cost savings should be conservative in nature, based on available metrics, tailored to the organization, and leverage knowledge gained in the past.
Obvious metrics that almost every organization can leverage include:
- What is the fully burdened hourly cost of senior and junior IT staff?
- Where is staff spending their time: is it on fixing production problems or on projects that will generate revenue for the business?
- What is the cost of downtime of mission-critical business applications?
- How much is being spent on penalties due to missed SLAs?
- What are typical hardware/software upgrade costs?
- How often are these upgrades occurring?
- What is the cost per MIP?
Others may include:
- Actual cost of a past data breach and potential costs associated with a future breach
- Code quality metrics
- Test coverage metrics
- Analysis of fault activity and associated production support costs
- Application performance trending statistics
- Hardware, software, and labor costs associated with using and maintaining full copies of production data versus using subsets for application testing
- Outsourcer costs associated with knowledge transfer or poorly documented applications
Often we find that capturing and leveraging metrics typically results from management-sponsored initiatives. Although metrics can be used for one time decision making, they are most effective when used as part of an IT organization’s normal business processes. This can help to monitor efficiencies and effectiveness in day-to-day development, deployment, and support of critical applications. The right metrics needed will be determined by the specific initiative and the management commitment to make the appropriate tools and data available to deliver those metrics.
How does the challenge of a retiring mainframe workforce impact costs? What can be done about it?
This is one of the more critical issues facing the mainframe environment. IT leadership needs to understand how much it is already costing them, with the business and technical knowledge of individuals who have years of expertise supporting their most critical legacy applications retiring. As they retire, will the remaining staff be able to keep those key applications running successfully? What does it cost when those key applications are down? How long will it take to train remaining staff to support them? Mainframe skills are in demand. How much will it cost to hire and train individuals to replace them? Will you need to hire contractors or outsourcers to take over?
One key to planning for turnover in the mainframe workforce is to provide tools that can assist and enable with knowledge transfer, that ensure applications are well documented, well tested, and that provide team leads with the metrics they need to accurately manage, scope, and staff projects.
Are there other savings that IT often overlooks?
Taking a look at standard operations with an eye for wasted resources or manpower can be very revealing. One area we find that is often overlooked is that of fault management. In working with our clients on fault management and tracking, we often see lengthy manual processes in place resulting in the generation of management reports which are neither timely nor accurate. Tools to automate the fault management process can help to quickly identify trends and applications which are causing more than their fare share of abends, and are not only expensive to maintain but are wasting CPU time in reruns.
Another area is that of test data management. Sites using full copies of production data for use in their test environments are using significant amounts of DASD, CPU, and labor. These costs can be reduced by implementing an automated test data manufacture process which results in using smaller, related extracts of production data.
What examples can you offer about Compuware products helping IT organizations to save money?
Our customers estimate that in the past six months, we have saved them more than $2 billion. We can say that with confidence because our customers share their cost-savings successes with us every day. These savings revolve around 3 key areas: resource efficiencies, vendor consolidation efficiencies, and process efficiencies. Here are just a few of the many examples.
A U.S. financial services firm was experiencing escalating monthly outsourcer bills resulting from spikes in CPU utilization. They leveraged Compuware’s Strobe product and services and in the first nine months, the customer saved more than $1 million. In the next year, they doubled that and in one month this year, they found one small change that saved them $800K.
Since including Compuware’s Code Coverage into their production promotion process, another U.S. finance organization hasn’t missed SLA’s due to an application failure in almost 3 years. Prior to acquiring Code Coverage, they were averaging $400K annually in SLA penalties.