The last time I checked, ROI stood for "return on investment." In sales speak, it's "spending money to make more money." As a manager, you know that your project approval depends on showing that the return will exceed the investment. Prove that you can do it in a year or less, and your project will most likely get funded. The corporation looks at the project as being "free" because you paid for it by pulling cost from other parts of your budget. The next year your project actually "makes" money for the corporation.
Back to reality
OK, now let's take a step back and look at ROI without the hype. In terms of your organization and its processes, ROI is about making changes in order to net more revenue. That's the "why." The "how" is through combination of reduced cycle time, increased capacity, reduced waste and/or lowered costs.
Hardware isn't so hard
When you're analyzing the purchase of a piece of technology, such as for a server upgrade, it's easy to get traction. Capacity and performance are already documented, and you have budget numbers for power, maintenance and floor space. Read the specs for the replacement, do some math, and if you come out ahead, it's easy to justify the purchase.
Service is a different beast
Because IT service delivery itself is so complex, coming up with a meaningful ROI prediction is itself difficult and complex. For a given case, your organization delivers a specific service against
Beware the first three letters of assumptions
The fact is, you can't manage or apply costs to what you can't measure. You won't know many of the answers unless you do some digging. And good digging goes deep. For example, in one customer environment we discovered 100 individual use cases related to or rolled up under 12 service delivery lifecycles. Many of the 100 use cases also related or overlapped with multiple lifecycles and had components that were dependent on more than one lifecycle as well. To roll everything up under one grand average and apply a magic ROI formula would have been impossible.
Benchmark and build a better machine
Better be prepared to ask a lot of questions in pursuit of ROI. As a tip, consider how industry applies time and motion studies to streamline assembly lines and keep manufactured goods costs as small as possible. You can do the same for IT service delivery. Find out what people actually do with their time. See who touches a given transaction, for how long, and how many times per day. Look for flawed handoffs and FTEs doing a fractional day's work. Following a whole process lifecycle multiple times will give you a good meaningful time and cost average to fulfillment and an understanding of how things work -- or don't work.
Heroes do their homework
This is tedious and time-consuming work, but there's no substitute for facts. You can come up with very accurate costs to deliver services, if you have the appetite and time to follow the lifecycles, resources, time and hard costs. If you don't, the good news is that there are plenty of good reengineering consultants out there who can get the facts for you. Their reports give you an opportunity to fix things instead of just automating faulty processes so they work faster. And with data and a process roadmap, you can challenge each software vendor to develop a smoke-free ROI. Not a hypothetical, typical ROI, but your organization's ROI -- one you can take to the CFO with confidence.
Greg Lenox, an expert in the processes, methods and practices of IT operations for customer support, help desk, asset management, inventory management, telecommunications management and change management, is president and CEO of Entuition Inc., a maker of operations management solutions in the infrastructure logistics marketplace. At Entuition and in previous positions, he led several major re-engineering projects. His clients included SunTrust Banks, Citibank, Target, GlaxoSmithKline, Baxter Healthcare, Nations Bank, Wachovia Bank, First Union Bank, BB&T, CCNB, CNA Insurance and others.
This was first published in March 2004