When businesses negotiate price with technology vendors or consider new projects internally, they frequently demand persuasive proof of a positive return on investment as a requirement to go forward. Often described as the time it takes to recover the cost of an investment with a resulting benefit, this kind of predictive ROI ahead of a large project is not an unsound idea. But such calculations always involve assumptions, some of which inevitably lose credibility on close examination. In such cases, ROI projections might be used to enforce an executive bias, or as a poker chip in a pricing negotiation - rather than as a good or bad reason for a company to make an investment.The point is that positive ROI is an outcome of good strategy, not a strategy unto itself. Though smart companies will continue to make smart investments, as scale and scope increase, predictive ROI loses predictability. In some cases the ROI fixation is simply inappropriate. How, for example, does one establish future ROI for a data warehouse project, a long-term investment from which most benefits will not be evident until years later?

Mark Smith, CEO and SVP of research at Ventana Research, has witnessed a good bit of hype in certain aspects of defining and justifying a return on investment for business intelligence. By the same token, Mark works with world-class companies that wrestle with such issues.

Whatever the investment, Smith says the starting point should be a business case backed by good, basic cost/benefit analysis. "ROI is the goal and the vision; it's not what you need to define to go through a corporate governance review," he says. "To justify an investment you have to say, 'this is going to help us address an issue.' What you have to do in the corporate governance review is justify your being, justify your investment."

In the case of BI, information is at the intersection of people and process. "You explain that somebody didn't have the right information and said yes or no, maybe making a wrong decision. As you define benefits and start to relate them back to the information, the question then becomes, 'what are the cost scenarios that include the software, hardware and personnel cost for both business and IT to achieve these requirements to reach our goals?'" It might be wise to also point out that software might solve only 25 percent of the problem, but is a requirement nonetheless.

Smith regularly comes across projects that have lost executive interest because of poor articulation skills around cost/benefit, leaving uncertain ROI arguments as a last-ditch appeal. "Many of our clients had been rejected by governance teams that weren't able to link an investment to benefits on the business side. All they had on the business side were people saying 'this is great,' but that doesn't pass muster for approval."

Understanding what a program is, what process is, having a business case aligned with a grasp of total cost of ownership, these things all have to precede empirical evidence of ROI. "The only way you know you can measure ROI is at the end of the project life cycle," Smith says. "If it's a three-year project and every year you are achieving benefits, you can calculate ROI based on the cost, the benefits and the post-mortem of what you achieved in the last 12 months." From Smith's observations, it's clear that ROI should be used at least as much as a tool for leveraging, adjusting and improving BI and data warehouse projects as it is a precursor for investment.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access