How CFOs can extract proof of IT profitability is a source of puzzlement. During the years of easy IT money, the established procedures for capital budgeting couldn't cope with a growing appetite for computerization. Now that companies have slammed on the brakes, CFOs are searching for new ways to harness IT.

The most elaborate scheme yet is the one the Office of Management and Budget conceived for constraining the more than $57 billion in annual federal IT spending (or $32,000 per employee). OMB efforts, conducted under the federal enterprise architecture and e-government strategies, are scripts for imposing order and uniformity on an IT landscape that comprises more than 100 independent agencies and includes more than 5,000 major projects. The OMB plan reflects the thinking of many CFOs and is therefore worthy of attention. It would do the following:

1. Impose a standard method for classifying IT expenses.
An elaborate business-reference model defines how to describe the business operations of the federal government. The purpose is to capture information about any duplication of efforts.

2. Prescribe a self-appraisal method for characterizing the performance of IT, with an emphasis on identifying system deficiencies.
The purpose is to identify targets where IT spending must be examined.

3. Require classification of governmentwide standard IT components to enable consolidation of application development.
The purpose is to set the stage for centralized procurement of software and services.

4. Direct IT organizations to conform with a technical reference model for reuse of technology.
The purpose is to establish favorable conditions for creating a shared infrastructure that would save money.

5. Require the submission of a "business case" for every project.
It's through the scrutiny of such business cases that the CFO intends to assert budgetary controls. OMB officials have already declared that "771 projects in FY04 budget (for $20.9B) are 'at risk' and will not be allowed to proceed." So how do OMB budget examiners sort out which of the more than 5,000 projects are at risk?

The so-called business-case forms have been conceived by auditors for budget examiners. For each project, the CIO of the organization must fill out a form containing 132 detailed questions about the best IT practices ever conceived by consultants. The checklist covers topics such as demonstrated fit with business strategy, linking to management plans, support of a modernization strategy, demonstration of low-risk acquisition methods, proofs of strong project management, closing of performance gaps, assurance of security over a project's life cycle, privacy protection, paperwork reduction, management of risk-adjusted life cycle costs -- and many others.

For good measure, the business-case forms require documented confirmation of compliance with a long list of regulatory and legislative measures.

To sort out which projects are at risk, the OMB follows a routine procedure. First, each item for every project is rated on a scale from 1 to 5.

Second, all of the scores are added up, regardless of their importance, though a poor score in a few selected areas (such as security) will automatically disqualify the entire project. If the scores fall below predefined levels, the project is classified as being at risk and scheduled for further examination. Projects with low scores won't be funded.

How effective is the OMB methodology in delivering the stated objectives of savings through sharing of IT resources? One way of judging that is to examine the OMB schedule of fiscal 2004 governmentwide sharing of IT. If you remove unique Homeland Security Department projects, you end up with only 0.21% of the total federal IT budget as benefiting from synergy through sharing. I guess that leaves the OMB -- like many CFOs -- with the hatchet-on-a-pole method of IT budget pruning and management.

Paul A. Strassmann (paul@strassmann.com) has been fighting budgetary amputations as a way of curing IT disabilities for over 40 years.