February 17, 2003 (Computerworld) --
To get a peek into the real world of return-on-investment analysis for IT projects, Computerworld talked toAmir Hartman andCraig LeGrande, co-founders of the management consultancy Mainstay Partners LLC in Redwood City, Calif. They've helped many companies measure the bottom-line impact of their technology investments, as well as identify productivity improvements. Which types of ROI models are you seeing used in corporate America?
Hartman: We're seeing a number of different things being used, and that perhaps is part of the problem. Internal rate of return, IRR, is pretty common. Return on investment capital, ROIC, is something we're seeing more and more, especially for large companies. Return on net assets, RONA, is something that big companies talk a lot about; the problem is translating projects to RONA is difficult.
LeGrande: Customers are using multiple ROI methodologies. So even within a company, you'll see some divisions use IRR, some use net present value. ... And there are challenges when you go from the corporate metrics -- many of the retailers use RONA, for example -- and try to drive that down to individual projects. It's hard to divide up assets in the projects. So one challenge is how to have a standard ROI process across the company.
Hartman: The companies actually using an ROI metric are few and far between, whether we're talking about RONA, ROIC or IRR. Using anything is a pretty decent thing, because our data shows that over half the companies don't use any metrics at all.
LeGrande: A lot of companies don't have a great process for educating their IT project managers on how to develop ROI. So each group is left to its own devices and comes up with different methodologies for ROI.
What's wrong with different divisions or project managers measuring ROI differently?
LeGrande: A lot of big companies are making the big capital-investment decisions centrally. And they want to be able to compare apples to apples. A lot of problems come up when a business case for one project is done completely differently from another business case. Some have different levels of detail or highlighted benefits. A CEO doesn't have a lot of time to get more information or redevelop a business case if he doesn't understand it. So having a standardized template for making those portfolio decisions is better for the senior leadership.
Hartman: We're not arguing that you should use only one of the acronyms, but that there needs to be a lexicon that's defined and well understood.
LeGrande: Like you hear the term total cost of ownership, but it means different things to different people. When senior executives try to weigh which investment really is better, it causes problems if they can't figure out whether TCO from Division A is the same as TCO from Division B. It makes it murkier.
A standard process and lexicon also help if you're comparing IT investments to other investments a company might make, such as building a factory.
Hartman: That's a very sophisticated way of doing things -- very few companies do that. But we definitely encourage companies to start looking at their IT investments as part of a portfolio of business investments, though they should take baby steps to get there.
LeGrande: The company can either look at its investment capital as one bucket, from which they can pull out the top investments, or they have two buckets, one for business capital investments and one for IT capital investments. In a down economy, the IT investments will be the first ones to go, because they're viewed as softer investments or less strategic.
There's a contrarian view that some IT projects are so strategic that doing ROI calculations is a waste. The project "just makes sense."
LeGrande: One of the problems with that is, without really defining what benefits you should see when, you never know whether that project is on track. Are we hitting the targets we thought we would hit? Just having some metrics or decision points helps the leadership figure out whether to keep investing.
Hartman: IT folks love to think that everything they do on the planet is earthshaking. Show me a project that, if you pushed it out two weeks to make time for a rigorous ROI, you'd kill the company. It's not like that.
LeGrande: There are small, two-week enhancements to systems that don't need an ROI calculation. So we help customers stratify the projects. The level of ROI rigor is based on size.
Getting to know Windows 7? Don't stop now: From speeding up taskbar thumbnails to reining in UAC, here are 20 ways to make Windows 7 act the way you want.
Sure, you could always use Linux as a desktop OS, but Corel Linux 1.0 was the first distro designed for ordinary users. It's been a long, strange trip since then.
See your link here
Getting to know Windows 7? Don't stop now: From speeding up taskbar thumbnails to reining in UAC, here are 20 ways to make Windows 7 act the way you want.
Is Motorola's new Droid good enough to vanquish iPhone envy? To find out, we took it on a 3-day trip.
Sure, you could always use Linux as a desktop OS, but Corel Linux 1.0 was the first distro designed for ordinary users. It's been a long, strange trip since then.
New touch-screen laptops from Fujitsu, HP and Lenovo take advantage of Microsoft Windows 7's touch-friendly infrastructure.
Get the latest news, reviews and more about Microsoft's newest desktop operating system.
General Mills, Genentech, San Diego Gas & Electric, University of Pennsylvania and Monsanto top the list.
Computerworld
