The cost squeeze is on. Automobile companies are racing to get slimmer. According to The New York Times, automakers are now ranked according to how few hours they spend building a car. For instance, General Motors claims to be the leader, with only 35.2 hours per vehicle. Chrysler puts in 37.42 hours. Ford takes 38.6 hours. They’re all keeping an eye on the efficiencies of Toyota, which assembles cars in Fremont, Calif., in only 21.92 hours while gaining market share.

Every company in the world is now engaged in a race to reduce costs. The hours-per-car statistics are revealing because they reflect the extent to which outsourcing takes place. GM looks good because it outsources more than Ford: 76.2% of revenue vs. 70.2%. You can always show an improvement in productivity by getting others to do your work. For instance, the current claim by GM that it reduced IT costs as a percentage of revenue doesn’t necessarily attest to the company’s improved efficiency. It only proves that GM has shrunk its value-added and therefore doesn’t need to spend more on IT.

Calculating the cost of GM’s 35.2 hours per vehicle raises a more interesting question. GM’s average labor cost, which includes managerial and executive compensation, is $33.30 per hour. That means GM puts $1,172 of labor into a car that it sells, on average, for $30,500. Even after paying for depreciation and taxes and making an allowance for relatively slim profits (another $6,087), this leaves $23,241 of purchases to be managed for greater efficiency.

Taking a purely intracompany view of the scope of IT would, however, be a mistake. The emphasis in global competition is now shifting from the costs of a company’s products to the consumer’s life-cycle total costs of ownership. In the case of a $30,500 automobile, that requires adding the costs of distribution and the customer’s five-year expenses for taxes, insurance, financing, registration fees, maintenance, repairs, oil and fuel. That raises the consumer’s cash costs of an average GM car to at least $73,600.

As I calculate the consumer’s life-cycle costs of $73,600, I find that IT can’t account for more than 6% of the total expenses. The costs of information management would be about 30% if added up for suppliers, manufacturing, management, dealers and the consumer’s ownership. As manufacturers shift costs to suppliers, as management shifts costs to manufacturing and as customers shop around for the best deal, every reshuffling of how money is spent ultimately affects all others. The cost of parts will influence maintenance and insurance costs. The cost of the car will influence taxes and depreciation.

In the global marketplace, all costs become interdependencies, often involving hundreds of businesses to deliver a single product. Therefore, the management of IT over the next few decades will be shaped by the need to extend the boundaries of corporate information systems from the traditional corporate-centric orientation to delivering consumer-centric services. For instance, in the case of an automobile, this will require setting up lifetime configuration management databases for preventive maintenance, repairs, parts, upgrades, warranties, financing, registration, insurance and fuel management.

The concept of global life-cycle support of products and services, as seen from the consumer standpoint, will become the dominant theme of IT management.This will require an architecture that assures data interoperability over the entire value chain. For instance, the reliability of parts must be traceable over a complex chain of ever-changing suppliers. Preventive maintenance records will have to travel as a vehicle passes from owner to owner. Accident records will have to be integrated with insurance and repair records.

Identical reasoning also applies to pharmaceuticals, medical services, food, banking, refrigeration and airplanes. The CIOs who succeed will extend the positioning of IT from the corporate back office (an overhead cost) to a customer life-cycle support view (a company’s core competency).

The current introspective, function-centric and process-centric systems architectures will have to be replaced by data-centric designs that assure interoperability of data that’s independent of technology, location or organization.

Evolution of Corporate Architectures

Implementing such a transition will be the challenge for all IT executives in the future. Meanwhile, the IT industry, the CIOs and all of the naysayer gurus should be put on notice that the best of IT is yet to come.

Paul A. Strassmann has served as a CIO long enough to recognize that the full potential of IT is ahead, not behind us.