acm - an acm publication


Beyond numbers

Ubiquity, Volume 2002 Issue September, September 1 - September 30, 2002 | BY Ubiquity staff 


Full citation in the ACM Digital Library

Martha Amram on the current economics of technology investment.

Martha Amram on the current economics of technology investment.

Martha Amram, author of the new book "Value Sweep: Mapping Corporate Growth Opportunities," is an independent consultant and adviser to Fortune 500 companies and the startups. She lives in Palo Alto, California.

UBIQUITY: Why would our readers -- most of whom are highly involved in information technology -- be well advised to read your book?

AMRAM: Many people in information technology (IT) must justify their IT project to finance types, those who control the budgets. Right now, they lack a way to communicate the value of the project, to identify the most important risks for business managers and to easily determine the monetary consequences of those risks.

The purpose of my book is to help close these communication gaps. It suggests simple quantitative frameworks, and the key features to describe to business managers. The book also highlights the importance of the storyline -- How will this project be successful?

For information technology, the storyline is really important. Information technology projects never take just one investment; they are done in stages. The projects also span technology and business issues. The storyline connects the many, many steps and the many audiences together into a single vision.

So one of the messages in the book is very important for information technology managers: IT valuation is not just the numbers, it's also the story.

UBIQUITY: What is particularly elusive about information technology compared to other kinds of investments?

AMRAM: Let me put the examples I use in my book on a continuum. At one end are well-defined valuation problems that use well-known and simple valuation tools. I did a valuation of the consulting firm KPMG that fits this description.

Information technology is at the other end of the continuum. Valuing information technology is not a well-defined problem and requires modifying more complex valuation tools.

There are two aspects that make valuing information technology projects hard. First, the value of information technology rests on the specifics about how other technologies are deployed in the corporate setting -- their vintage, their maintenance and so on . These factors vary greatly from one corporate environment to the next, and can significantly change the value of an information technology project. Second, information technology is not a stand-alone investment. It could be part of a product development initiative, it could enable financial controls, or require investment in some non-information technology. For these reasons, putting numbers on the benefits of an information technology project can be very difficult -- what is in the project is not always a well defined.

I take a pretty sharp point of view in the book, arguing that we really can't precisely value information technology projects. The best valuation tools simply do a crude sorting of projects by value.

UBIQUITY: Do you see fundamental differences between types of information technology investment?

AMRAM: Yes. There are basically two types of information technology projects. Those that reduce costs, which can be measured by an ROI type model, and those that create options and need other valuation tools.

Currently, with our slow economic climate, only projects that reduce costs are being approved. Hence the current emphasis on ROI as the project selection criteria. But in the bubble, the ROI criteria were thrown out. Not in a fit of foolishness, but because the tool fails to capture the upside potential or option value of many information technology projects. For example, building a Website creates the option of gaining online customers. That potential gain can't be measured by ROI.

UBIQUITY: I think everybody is probably familiar with real options at some level. But why don't you tell us what real options means?

AMRAM: Real options is the application of financial option pricing mathematical models and ways of thinking to real or non-financial assets. For example, mathematical option-pricing models are used to calculate the reported employee stock option value in financial statements.

Option pricing models have proven useful when applied to real assets. For example, real options thinking has been used to price contract terms in DRAM contracts. It has also been used to think through e-commerce initiatives because real options thinking is so useful in characterizing stages of investment, particularly when the decision to continue a project depends on a future uncertain outcome.

Because of its origins, real options links project valuation to financial market pricing. For example, real options shows how the value of a startup in the semiconductor industry depends on the current stock market valuation of mature semiconductor companies.

UBIQUITY: You mentioned the different economic moods -- 1999 was very different than today, 2002. Would you expect a whole new environment in 2003? How often do these moods change?

AMRAM: My own expertise is about how the economic moods affect valuation, but not on the timing of the business cycle. From my valuation perspective, I see two affects of the shifting economic climate.

First, we're entering a period in which firms are running low on cash. This means that there are no funds to execute growth strategies, so despite the quality of a growth option, valuations will fall because of a smaller probability of executing all the steps needed for completion and profitability. There will also be continued dominance of cost-cutting, ROI-proven information technology investments.

Second, the shift we've gone through from 1999 to today has been dramatic and huge. This particular history makes today seem even more gloomy. Here in Silicon Valley we vividly remember zero percent office vacancy, the difficulties in attracting engineers of any quality, and the easy venture capital money. Now it is high vacancy rates, no engineering job growth and a stall in venture capital financing. This sharp contrast affects our emotions and dampens valuations. It will take another year or two to work through this particular path of history and get back to "normal" conditions.

UBIQUITY: Let me read a sentence from your book, page 194. "Most e-commerce initiatives in the '90s were selected and chosen by the optimism of strategic thinking alone, without value-based decision making. In many cases, there was also a staged-growth mentality: 'We won't make money off this investment today but the two stages down the road we'll hit the big payoff.' With the collapse of the Internet sector and the stock market much of this thinking has been distorted. So where is the value in Internet technology? It's time for a review."

Let's do that review.

AMRAM: At first, it seemed that the Internet promised so much more than just the adoption of a new technology. It promised to disrupt established industries and to break established distribution channels.

As it turned out, the promise was bounded and the value of Internet technology was capped by the required complementary investments. To adopt a new business model based on Internet technology, we needed a change in business processes. This required asking people to change their work habits, their workflows and their organizational processes. Organizations have a natural pace of change and a maximum pace of successful change. After a while, companies could not absorb any more Internet technology or disruption.

And for a while companies like Enron were held up as models of new organizational forms. For example, Enron had an internal labor market that allowed people to voluntarily migrate to new projects at the same job level if they wanted. The company used this employee freedom as a way to sort out good project ideas from bad -- employees would vote with their feet.

Adopting this type of flexibility requires a huge organizational change for most companies. And so the adoption of the Internet technologies enabling this change was bounded and slowed. Further, as the economy slowed, companies demanded a clear payback to change. But organizational change is never completed in one shot, and the early stages often lack a defined result. The slowing economic climate accelerated the deferral of further Internet technology investments.

Companies have gone back to those projects that have near-term cash flow, which are not particularly innovative, but continue to drive cost. Internet technologies are being adopted only if they are part of this solution and don't require much change in anything else.

UBIQUITY: Are you saying that now that Enron has gone from a model to a ghost that its most prominent ideas are discredited?

AMRAM: No, I think my point here stands regardless of their accounting policies. For example, other energy companies such as British Petroleum and Shell Oil have experimented with organizational change as well. But most organizational change efforts are in hibernation.

UBIQUITY: One of the many interesting things about your book is in the epilogue where you talk about the personal qualities of leaders. Why don't you talk us through some of those qualities.

AMRAM: Valuation tools are simply descriptions. The results are based on the assumption that the growth project has successful leadership, so in my book I think about the required leadership qualities. Experienced investors understand that "The Guy" who leads the project is key. For example, venture capitalists often say they invest in the team, not the idea. Their thinking is that the idea will change, and a good team can navigate the change.

We often think of leaders as having boundless energy, but I argue that a bit of laziness is needed as well. Too much energy leads to unfocused efforts.

We often think of leaders as charismatic extroverts, but I argue that a bit of an introverted nature is needed as well. Too much talk and too little independent thinking leads to a lack of an innovative strategic vision.

We often think that leaders need to delegate, but I argue they need to roll up their sleeves and get to work. Leading a growth project stretches resources pretty thin. Sometimes if the leader doesn't do the work it simply won't get done.

UBIQUITY: Herbert Simon, the Nobel Prize winner, used the term "satisfice" to describe this behavior. For example, you describe doing the business plan on one page instead of on a mega-spreadsheet with numerous tabs and scenarios.

AMRAM: Exactly. Focusing at the right level is critical in a growth project. Growth projects typically involve a huge amount of detail, which can absorb all the leader's energy. Yet, the leader needs to drive an organization and lead a team of people, which also requires an extraordinary amount of mental space, time and energy.

So unless the leader learns how to satisfice, to do just the amount of work on each task required for a satisfying outcome, the leader will quickly burn out. Further, with too much detail, the leader's brain is cluttered and blocked from noticing the big, yet subtle, shifts in the competitive landscape.

UBIQUITY: Let's finish on a note of advice. What would you advise someone, maybe even a young person who has an entrepreneurial spirit and an idea or two? Would you advise him or her to hold it for a while or would you say, "Go West, young man?"

AMRAM: There are two groups of thought on that one. One group says, in any climate the best thing a young person can do is invest in him or herself. For most, that is getting experience before running with the entrepreneurial idea. On the back of my book is a quote from Ted Dintersmith, a venture capitalist at Charles River Ventures. Many years ago Ted told a university class of mine to choose their first job well. Ted's argument is that your first employer's work habits become your default work habits -- make sure they're good ones. Default work habits are key, as the entrepreneurial job is so unstructured.

Another group argues that now is a great time to start a company -- no one else is doing it! Everything is much cheaper and in a year or so, you'll be in a class by yourself.

Recently, I've been CEO of a software startup and my own take is that successful startups have customers. Right now it is very hard to get customers for high-tech products. When every potential customer says "no thanks," it is difficult for a startup to get meaningful product feedback. Thus, in this climate a startup runs a huge risk of building a product that no one wants. So perhaps it would be wise to wait a year or so, meanwhile picking up good work habits at a more established, well-run company.


[Home]   [About Ubiquity]   [The Editors]  

Ubiquity welcomes the submissions of articles from everyone interested in the future of information technology. Everything published in Ubiquity is copyrighted ©2002 by the ACM and the individual authors.


Leave this field empty