TCO vs. ROI vs. Payback Time: Is One Better?

In the best and worst of times for IT budgeting, internal debates and analysis within banks and other financial institutions will invariably get around to some financial justification for incremental new spending, evaluating new projects, or looking at contracts up for renewal in one or two years. Over the past three decades there have been several methodologies that bank finance types and third party consultants have used to perform this type of analysis. None of the alternatives are perfect and none always gets the right answer for a bank's circumstance. The most common alternatives are: Total Cost of Ownership; Return on Investment; and Payback Time. Is one of these the best? Should the line of business executives and IT folks care? What about the finance?Total Cost of Ownership (TCO): This is the most complex alternative and can take quite an effort to complete for larger projects. Ultimately, the total cost of an initiative has to be captured - everything from hardware, software, outside services, and all internal costs. This alternative tends to be reserved for the very biggest projects that will have an extended implementation life and long term use value to the institution. For example, a massive core banking conversion would be a candidate for a TCO analysis. More than once, the end result of such a TCO analysis contributed to delaying, substantially modifying, or even canceling a big project due to the size of the TCO. Banks that end up in this predicament usually go back to the drawing board and revisit priorities and other approaches that will do less for the business with less money and be better than nothing at all. I can't say I am a big fan of TCO - cost overkill can too easily creep into the analysis - but the bank still needs the analysis. Business and IT should both know what goes into TCO. Sharing TCO ownership between business, IT and finance is the only option to build consensus or limit criticism.

Return on Investment (ROI): This alternative used to be the most common, but has diminished in status on both ends of the IT investment spectrum (big projects use more TCO) and the most smaller projects have migrated to using Payback Time. ROI starts with the IT investment spend. Against the investment are the returns, based on what business and IT expected to realize from the investment. Finance usually owns this analysis. ROI would be delivered against a metric that the institution likes to use, often tied to the cost of capital to the bank. If a 6% ROI was needed over a five year horizon for a $10 million investment, then a simplified version of the analysis would require a net "return" of $3 million over the five year period (6% of $10 million = $600,000 annualized). Quantifying the "return" became an "assumption" game for some banks. I consider ROI to be a weak alternative.

Payback Time: Top management likes this alternative. The bank spends money and gets it all back with savings (usually the dominant source of payback) and, rarely, with identifiable new sources of revenue, such as fee income or discrete new, organic business. The shorter the time period for payback, the faster the approval. Verifying payback must be reasonable and reported. In tough times, six to nine months can be the hurdle. During better times, the horizon may stretch to 12 to 18 months. Simply put, management buys into a short term bump in spending that is quickly recovered in the near term with lower spending that leaves the bank ahead of the game for a long duration. Often used on smaller projects that have short implementation time periods, payback time is not a good fit for new, big projects or long lead time conversions to gain cost savings. There is no good business reason to exclude this alternative from the analytical tool set.

What alternative will work best for your institution? Or, does it make sense to use two, or all three, of the alternatives and pick the one that fits a particular project the best? The analytical process is not pure science, so I believe there is sound business logic to using more than one alternative (i.e., Payback Time, TCO) to get a best fit with the particular project or IT initiative. Hopefully the finance team will not be locked down on just one alternative and forbid the others.

Bill Bradway, founder and managing director of Bradway Research LLC, analyzes the business strategies and IT investments of US banks and credit unions.