Taking the Trillion-Dollar Figures out of the Costs of IT Failure

Dennis Byron

There were two highly visible blog posts during the week of September 28 on the cost of IT failures to the geo-economy. But just as all politics is local, even in geopolitics (see President Obama's whirlwind trip in and out of the Copenhagen International Olympic Committee meeting October 1/2), all the costs of IT failures are local. They hit your IT budget and your enterprise's bottom line first. Then each of your individual "IT failures" add up to a worldwide total.


"IT failures" is in quotes because before one can even begin to add up the cost of IT failures, failure must be defined. As illustrated in the comments section of the Michael Krigsman ZDNet post on September 29, "Annual cost of IT failure: $6.2 trillion," there is really no good definition. Krigsman is bringing up the subject in reaction to a post by Roger Sessions on his blog, in which Roger posits a calculation of the geo-economic cost of IT failures. I like Roger's approach primarily because his formula was coincidentally similar to one I posited trying to answer a similar question on my IT Investment Research blog on September 29, "Spend More on IT To Save a Trillion Dollars on Open Source".


Roger definitely asks the right questions, but I think part of his methodology may be flawed. The major issues I have with his methodology also apply to how you analyze your individual results within your IT department and enterprise. The issues are:

  • Should you count all IT costs including overhead and personnel or just the cost of purchased IT goods and services in estimating the cost of an IT project, a must-have before you can calculate the cost of an IT project's failure... should it fail. Roger says the latter. I say the former. You say tomahtoh, I say tomaytoe, except that when dealing with the geo-economy, we're talking trillions of dollars of difference in the final number. Unless you do IT for the U.S. government, you don't have that problem. You can get a good fix on all your costs.
  • How should we estimate the worldwide missed opportunity cost (often called the indirect cost/benefit of a project or conversely the expected return on investment--ROI). Again, you don't have that problem; you know your expected ROI. We analysts wouldn't struggle with this part of the equation if all of you reading this post and all your peers worldwide would just e-mail Roger and me your ROI numbers for every IT project you will spend money on this year. (Or all of you could just sue SAP--seeDumpster Diving: Rules on How Not to Acquire Enterprise Software --and we could find out via court filings.) Since neither is likely to happen, we need a proxy for the theoretical worlwide missed opportunity number.


Roger chose the aggregate gross domestic product (GDP) of all the countries in the world as reported on by the IMF or World Bank or someone as the proxy for the aggregate expected ROI/missed opportunity cost of the world's IT projects in process. I think he did it that way because he was only counting purchased IT goods and services as a cost as indicated in the first bullet above (or did it happen vice versa?).


But that's not the way you do the ROI on an IT project. You don't expect to save money out of your ongoing IT-sysadmin outsourcing expense when you are implementing a new purchasing application. You don't expect to reduce your firm's electricity bill by adding a new e-commerce feature to your Web site. Furthermore, many IT projects such as payroll processing are simply a cost of doing business, or involve meeting a government regulation such as Basel II or SOX, and so forth. There is no ROI at all expected of either (unless you want to calculate the fines you'll get from the various states for not paying your employees on time or from the SEC for not filing shareholder data in a timely manner).


So how do you calculate ROI on proposed or in-process IT projects? I don't pretend to have the answer yet but I'm uncomfortable with the trillion-dollar estimates floating around the blogosphere. Years of studying market dynamics tell me the market is not that inefficient; you cannot all be spending so much money (about 5 percent of enterprise revenue as explained on my personal blog) on something that isn't working out for you.


So help us help you calculate the cost of IT project failure and we can help you avoid such failures altogether. Send in your ideas on calculating ROI.

Add Comment      Leave a comment on this blog post
Oct 2, 2009 10:53 AM Michael Krigsman Michael Krigsman  says:

Thanks so much for focusing attention on this important topic. As you said, there is no easy way to calculate the numbers, however, we need to find more precision.

Regardless of the exact number, it's clear that worldwide IT failure is a huge problem. Accurate sizing of the problem is a key step to raise awareness of the significance of IT failure.

Some observers have commented to me that the number in my post (Roger's number) couldn't possibly be correct. Personal opinions are fine, but we need serious researchers to dig into the assumptions and math. That's how to advance this issue in a substantive manner.

Oct 2, 2009 11:10 AM Roger Sessions Roger Sessions  says:

Thanks for helping to bring this problem to the forefront.

My goal on choosing the numbers I chose was to be as conservative as possible. I don't mind people saying that I under calculated, but it hurts the cause when people think I over calculated.

I chose GDP because it is both available and reliable. And it relates to the report by World Technology and Services Alliance that gives spend rate on IT as a fraction of GDP.

I agree with you that it would be more accurate to include personnel costs, however that number is not readily available. In any case, it only makes the final numbers worse.

I agree with you as well that more accurate numbers on ROI would be helpful, but again, those numbers are not available I feel my numbers are, if anything, conservative.

It isn't really important if we are wasting $500 billion per month on IT failures (based on my conservative calculations) or if the real number should be $1 trillion per month (using less conservative numbers). The real point is this: This is a huge problem and we need to start focusing on understanding the causes and cures.

Those companies that do tackle this problem successfully have a huge competitive advantage over those that don't.

By the way, if you are interested in what I think is responsible for these failures, see my blog at http://bit.ly/M0Z55.

- Roger Sessions


Post a comment





(Maximum characters: 1200). You have 1200 characters left.



Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.