Let's recap a few things about ROI models that this story nicely illustrates.

It's all about the pronouns. This salesman (actually happens to be a male in this example) was basically trying to use his ROI model to justify the purchase of his product to enable him to meet his short-term business goals. Our business, our goals, our priorities, our processes and our risks were all noticeably absent from consideration (note that our costs were specified, though not emphasized).

A couple of things to watch out for here. Many such ROI models purport to be built on your data (for things like labor costs, interest rates and forecasted workload). Dig in further here. How solid is this data? Who within your company has vetted it? How much variability is embedded into it? Moreover, the logic about how this data is used is typically not at all transparent, which leads to our next key point:

What is the model assuming? Most ROI models make extravagant claims about cost savings that are based on very specific logical assumptions. In particular (classic phenomenon), ROI models typically assume that optimization based on forecasted values will be implemented as if this demand is the ordained truth. In contrast, everyone involved with a particular process will be looking closely at their own incentives ("management really looks hard at my labor-to-sales ratio," "I take a huge amount of flack if my on-time percentage is too low," etc.) and will do everything possible to manage their piece of the puzzle, and their localized risk, accordingly. Frankly, given the quality of 99 percent of what people in the business world call "forecasts," you would probably do the same thing if you were in their shoes.

This is a basic, primal phenomenon. Gene Woolsey has talked about this for roughly the last half century (see Woolsey, R. E. D., "Real World Operations Research: The Woolsey Papers," edited by R. L. Hewitt, Lionheart Publishing, 2003). Yet nearly every ROI model I've ever seen ignores it, for the simple reason that it's hard to model well.

Some key questions. Suppose for the moment that with diligent questioning you have managed to understand the crux of the ROI model that has been created. Then you have to ask yourself some fundamental questions:

• How confident are you about the savings and the time needed to achieve them? Because the costs are almost always up-front and definite, and whoever funds this investment will notice that immediately.

• Do you believe the underlying conceptual model? You will have to present it "up" in your organization, both to your management and to finance. They will have a much shorter attention span for the details than you do, so you need to be able to present the core story succinctly. If you don't believe it, you can be sure that they will not either.

• What resources and commitment are needed from which key people or groups to achieve the results that the ROI model claims? Do these people understand what is expected of them? What are their personal motivations to see this project succeed (or fail)? Are any costs associated with these commitments reflected in the ROI model itself?

Skeptics will abound. If you're not skeptical about the ROI model in front of you, either because you believe everything you're told or because you feel that you've dug into it and understand the answers to the above questions, great! But, remember that others who have less invested in the results that the model claims, most notably executives, are likely to be far more suspicious.

There are a number of reasons for this suspicion — overlooked costs, suspicious cause-and-effect relationships, risks associated with execution and all of the things that we've already talked about. Beyond all that, there's also one that we OR types never even think of: Simply saving money, even in the recent economic downturn, is not automatically at the top of the executive agenda. Peter Mayer's recent survey [Meyer, P., 2002, "Is Money the Right Pitch?" Business and Economic Review, Vol. 20, No. 3, pgs. 20-22] reinforces well-established ideas (Woolsey again!) that there are in fact different questions that most leaders focus on, such as identifying areas for growth, simplifying business processes and eliminating absurd practices while keeping the machine moving.

To be fair, ROI models have some real virtues, even when presented by salespeople. They are an instrument for structured data collection and business process review, which is often instructive on its own. When your own business processes are far more complex than what the model includes, it may be an opportunity to consider why you do things the way that you do. ROI model results may enable you, through judicious questioning, to learn something about how your competitors are handling similar situations.

But here's the kicker: The people who foist the ROI model upon you are almost never around to deal with the complexities of trying to achieve the results that their model promised would magically appear. The people that you presented the results to almost always are.



Vijay Mehrotra (drvijay@sfsu.edu) is a faculty member in the Decisions Sciences Group in the College of Business at San Francisco State University and an operations management consultant.

Table of Contents

OR/MS Today Home Page


OR/MS Today copyright © 2004 by the Institute for Operations Research and the Management Sciences. All rights reserved.


Lionheart Publishing, Inc.
506 Roswell Rd., Suite 220, Marietta, GA 30060 USA
Phone: 770-431-0867 | Fax: 770-432-6969
E-mail: lpi@lionhrtpub.com
URL: http://www.lionhrtpub.com


Web Site © Copyright 2004 by Lionheart Publishing, Inc. All rights reserved.