Five Keys to Establishing Better IT MetricsBy Samuel Greengard
Over the last decade, CIOs have faced growing challenges related to managing enterprise information technology and ensuring that systems deliver maximum results and ROI. At the center of this complex equation: IT metrics.
The ability to gauge costs, speed, efficiency, performance and overall productivity is critical. Here are five factors that CIOs should focus on:
The wrong metrics wreak havoc. The irony of metrics is that it's possible to hit every goal spot on and yet fail as an IT organization and as a business. If an organization winds up measuring the wrong things, projects can sputter and results can fall short of expectations.
"You have to take a balanced approach," says Merim Becirovic, senior director of business operations at Accenture's High Performance IT Organization. "The metrics structure must be circular so that you're open to feedback from the business at all times. You won't always get it right, but you drastically improve the odds for success."
All metrics must ultimately connect to customer performance. While it's wise to look at what other organizations are doing and what measures and tools are commonly used within an industry, "It's critical to look inside your own organization and identify critical metrics," Becirovic says.
What's more, these metrics increasingly touch multiple parts of the business. Metrics should connect to a clear execution plan, and the IT organization must ensure that the resources are available to achieve the desired results. In the final analysis, "It's important to monitor extensively and roll up the data to know whether the organization is on track," he notes.
Metrics are more than numbers. The best metrics provide a glance into the past but also offer a glimpse into the future. They allow a business and an IT organization to understand where exactly it is and what direction it must achieve desired results.
"Metrics should translate into definable actions that help build the business," says Gene Siciliano, author of Finance for Non-financial Managers and president of Western Management Associates, a Los Angeles financial management consulting firm. What's more, people must be able to use the data. This means presenting information in a form that people understand—including charts and graphs—and educating staff on how to put information to full use through dashboards, emails, newsletters and other tools.
Every project requires customized metrics. It's not unusual for larger organizations to have hundreds of IT initiatives taking place at any given moment. While it's tempting to streamline processes, it's important to avoid cutting corners by measuring too few things—or applying general templates.
Many organizations rely too heavily on common metrics, such as whether a project is completed on time and on budget. This occurs because these metrics are easy for people to understand.
However, "You have to use metrics that match specific projects," Becirovic explains. "You really have to understand the challenges the organization faces by deploying a particular system or application and how the metrics impact the organization, customers and other end users."
Metrics are a moving target. As IT organizations evolve from a more traditional structure and a waterfall approach to an agile framework, metrics change. Things that previously mattered fade into the background while other factors become critical to success.
But simply looking backward won't help since the metrics and lessons learned in a waterfall organization are distinctly different from those in an agile organization. This may require IT to blow up existing metrics and completely rethink what matters to the organization.
Then, over the ensuing 12 to 18 months, a CIO can begin to compare, contrast and evaluate results in order to better understand which metrics matter most. At the same time, it's important to avoid analysis paralysis and periodically review metrics to ensure that they remain valid and valuable.
Samuel Greengard, a contributing writer for CIO Insight, writes about business, technology and other topics. His forthcoming book, The Internet of Things (MIT Press), will be released in the spring of 2015.