Coke ovens over computers.
At Bethlehem Steel, it's painful to tear away money slated for rebuilding a coke oven to buy more computers. The steel industry is a very competitive, very capital-intensive business, and a couple of years ago we were beginning to realize that we needed to focus on what gives our company its competitive edge -- our expertise in steel-making.
So, in late 1991, I met with Bethlehem Steel's senior vice president of operations (he's now our president and COO) to discuss a big concern of ours: the company's inability to leverage information technology (IT). We knew that many companies, especially manufacturers, find keeping up with technology very difficult, and we were among them. But we also knew that that's how a firm makes its technology investment cost effective. You simply can't install costly information systems that aren't contributing to your business. We decided that outsourcing our management information systems (MIS) function may be the answer for us.
THE STARTING GATE
But first we needed to educate ourselves about the capabilities of an outsourcing vendor. To do this, in December of 1991 our chief counsel (and ultimately our new CEO) arranged for a confidential meeting between a leading outsourcing firm and me. We realized that, at this stage, even the suggestion of outsourcing could have created serious employee morale problems within the lower levels of the firm, so we didn't openly pursue the project.
I subsequently informed the COO of what I had learned about the depth and scope of these services. He in turn discussed the opportunity with our CEO. (The CEO at the time has since retired.) Since the CEO was in favor of our continuing to explore the prospect of outsourcing, following that meeting the senior vice president of operations and I revisited the outsourcer for more education.
Finally, in March of 1992, having gathered sufficient information about the outsourcing business, we were ready to address the executive staff at Bethlehem Steel. Our proposal to the executives was straightforward: Based upon our understanding of the capabilities of outsourcing vendors, Bethlehem Steel should very seriously consider entering into a partnership with such a vendor whose core competency is IT.
We anticipated the primary question Bethlehem Steel executives would ask: Why should we outsource now? We were prepared to answer with four major reasons: First, we had outdated hardware. It wasn't junk, but it was old. We were two generations behind in a rapidly developing technological environment. Second, our software systems also were outdated, some developed as early as the 1970s. Third, technology is costly. We knew we could spend a lot of money on IT without having much to show for it. Fourth -- and our driving concern -- today's technology is so dynamic that we, as a steel company, can't hire and retain enough smart people to keep up with the changes, even to know what's available from month to month. We were totally at the mercy of our vendors to tell us what to do, and we didn't like being in that position.
From this list of problems we faced with our MIS function, we developed a list of objectives to guide us as we evaluated outsourcing candidates. Here's what we wanted:
* Access to all technologies -- With an outsourcer, we would be able to tap into any technology the vendor offers, no longer being limited to what Bethlehem Steel had in-house.
* Access to critical skills -- We'd also be able to make use of the outsourcer's skills bank, since a vendor typically employs thousands of IT experts.
* Capital avoidance -- Like many companies, we'd rather use our capital to improve areas other than IT. When you're in the steel business, spending money on rebuilding a coke oven always feels better than spending money on a computer system.
* Reduced costs -- We wanted to save money by no longer purchasing and maintaining hardware and software -- and no longer employing a staff of technology professionals, which typically makes up the biggest part of a company's MIS budget.
* Reduced training expenses -- The fewer IT responsibilities we have, the less training we'll need to provide to stay on the cutting edge.
* The opportunity to manage our business, not technology.
* Flexibility and incrementalism -- Although we had some flexibility in application development through the use of contractors, we weren't totally satisfied with the arrangements.
More important, we frequently faced situations in which a small, incremental requirement in data center service (for example, something involving the central processing unit or direct access storage device) would trigger a substantial capital requirement for something like a completely new computer or a controller.
* Predictable costs -- We wanted to be able to more easily project our IT expenses. When you have a contract with an outsourcing vendor that spans five or 10 years, you can do that.
* Reduced risk -- We were trying to reduce the risk associated with selecting the wrong hardware and being stuck with it. Likewise, on the development side, we wanted to reduce the incidence of large, costly and time-consuming project overruns.
* Improved quality and speed -- We wanted faster development of applications that, when installed, performed as we expected without substantial rework and modifications.
* A competitive advantage -- I believe that no single act will result in a permanent competitive advantage, since one way or another your competition can always copy whatever you've done. But we figured if we could leverage IT faster than our competitors we could gain an initial advantage that would then be up to us to maintain.
* Better control -- In our judgment, traditional management strategies simply won't work with IT. The rate of technological change and the inherent IT cultural issues require managerial strategies different from those used for the rest of the organization.
* A foundation for change -- At Bethlehem, we've had the gurus describe the aggressive objectives of business process re-engineering, and we're now pursuing several process simplifications. An underlying assumption is that IT will usually play a key role in developing systems to facilitate these changes to the business processes. Therefore, a partnership with an IT services company could allow us to develop systems rapidly and cost effectively.
BULLETS TO BITE
As we formed our objectives, we also identified the four risks of establishing a relationship with an outsourcer: vendor failure, increased costs, loss of control and loss of skills.
We knew that the number one risk for any company in an outsourcing arrangement is vendor failure. Say we signed a 10-year agreement with an outsourcer and, before the ink was dry, the wheels started to come off the wagon. Then what would we do? And what would happen if, two years after we enter into an outsourcing agreement, we look at what we're actually spending in vendor and other fees and find our total costs have increased, not decreased? Plus, when we transfer all of our MIS people to an outsourcer, that represents a huge loss of control and loss of skills for Bethlehem Steel. We asked ourselves, what if the plan goes awry? How do we get those people back? How will we recoup?
We did believe that vendor failure could be a near catastrophe. We pragmatically discounted vendors' claims that they would "make us whole" if the agreement fell apart. But we clearly informed our executives that, in our case, bringing IT back into Bethlehem Steel was not a likely possibility. Our only course of action would be to find another vendor. These assumptions were key drivers in selecting a vendor and crafting a contract in which the potential for failure was minimal.
In March of 1992, after the presentation to the executive staff, we formed a team to begin studying outsourcing in depth. Dubbed the "Dutch Group," the team was sworn to secrecy and functioned from an off-site command center, in a totally confidential manner, for almost eight months. The membership consisted of the top management of our existing IS department, a process control expert from our research department, and me. Additional personnel from accounting, finance, human resources and law participated as needed.
Because we knew that outsourcing agreements are frequently arranged at high corporate levels with a handshake before any of the technical issues have been resolved, we established one procedural requirement, with the support of Bethlehem Steel's executive staff: Each vendor was warned that any "end-around" attempt to bypass the Dutch Group and arrange a deal with the executives would be cause for that vendor's immediate disqualification. Our strategy was to have the Dutch Group question, challenge and, when possible, physically confirm every aspect of the vendors' capabilities. We wanted no marketing glitz from the vendors until our people were satisfied that each outsourcer could, in fact, deliver what it claimed.
Through April and May, the senior vice president of operations and I continued to visit outsourcing vendors. We considered seven vendors, but we quickly narrowed the field to two because of Bethlehem Steel's particular interests: We wanted to achieve seamless computing capabilities between our substantial installed base of process control computers on our factory floors and our extensive investment in business-oriented computing systems. Conceptually, we were driving toward an environment in which we would minimize the "this-is-a-data-center-application" and "that-is-a-process-control-application" attitudes. We wanted to do the right thing in the right place in the most cost-effective manner possible.
From May through August, we held a continual dialogue with these outsourcer candidates. The one rule we established for our discussions? No surprises. We shared absolutely every bit of information we could find that the vendors wanted to see. Our reasoning was that the less we shared with them, the higher was the risk that we wouldn't get the IT support we needed. The higher the risk, the higher the cost.
Many companies disclose nothing, because their executives fear the outsourcing candidates will use the information to gain some advantage over them. It doesn't work that way. We talked through every step -- and didn't limit the exchange to IT issues. Having signed mutually binding secrecy agreements, we were free to share our operating and marketing strategies as well as our key strengths and weaknesses as we saw them.
Of course, we made numerous visits to the competing vendors' data centers and even to their existing customers. Likewise, the vendors' representatives visited Bethlehem Steel's facilities to get a better understanding of our operation.
Eventually, our executives got involved in the decision-making process. This included our chairman, president and CEO; our senior vice president, secretary and general counsel; our senior vice president of operations; our executive vice president on the commercial side; our senior vice president and CFO; and another senior vice president.
They visited the facilities of the two top outsourcing candidates and participated in sit-down interviews. The interviews were one-on-one discussions between Bethlehem Steel executives and the key people from each potential outsourcing partner. These discussions helped us better understand the outsourcers and helped the outsourcers better understand the steel business and where Bethlehem Steel fits into that business. We admit we're in a very difficult spot, and the vendors rightfully want to know how the future looks to us and what our plans are.
Then, during July and August, we developed our request for proposals. Our target was to receive final proposals on September 1 from the two top competing vendors. And, on September 1, we did receive a proposal from each, covering the entire scope of the job we had outlined over the past year.
We scheduled a day for each potential partner to make a formal presentation to the executive staff, one on September 10 and one on September 15, and we moved the process to a neutral site in New York City. For about six weeks, we negotiated simultaneously with both vendors, ironing out countless details, technical as well as financial.
By November, we had narrowed our major negotiation issues to these items:
* Exclusivity -- We're in a very competitive business. Because we have capabilities we've developed within our steel mills that give us a competitive advantage, we want to make sure these capabilities don't inadvertently find their way to our competitors. So we ask our outsourcer not to work for any other company in the steel industry.
* Delivery -- We wanted continuous improvement in performance levels throughout the life of the agreement for numerous items.
* Pricing -- We wanted a flexible pricing structure, because we're convinced that, if you want to evolve in this world of technology, you need a partner who's willing to develop a pricing structure that won't inhibit you from doing the right thing in the right place.
* New Projects -- We knew at this point that the savings on the basic services were going to be substantial, and we always had intended to reinvest a significant portion of the savings into key quality and cost projects that we wished to expedite.
* Inflation Protection -- In a 10-year deal, many millions of dollars are associated with this unknowable inflation rate. We wanted to share this risk in an equitable manner.
* Ease of Dealings -- The scope of our particular deal encompassed a variety of technical and financial trade-offs that at times made the negotiations both complex and tedious.
* Teaming Provisions -- One of the proposals included a cooperative effort involving three outstanding companies: IBM, AT&T and Digital. This brought up some additional issues about which partner would be doing what.
* Ability to Close -- Our negotiating team was very highly empowered to make the many trade-offs within and between technical and financial issues. Whenever the vendors' representatives weren't similarly empowered, the process ground to a halt.
Because we were so methodical about our selection process, we called on many different people to help us in the evaluation.
One key to making a sound outsourcing decision, we believe, is that Bethlehem Steel's top people were directly involved in and committed to the effort. Our CEO made many visits to the offices of outsourcer candidates. In fact, everyone on the executive staff spent a lot of time on the project. We also involved the head of internal information services and two other people, including me, representing process control and the interests of Bethlehem Steel's plants.
To their credit, the executives at Bethlehem Steel didn't pretend they knew about the technology during the process. They expected the people within our organization who did know about it to make honest, objective evaluations and to counsel them on the pros and cons of outsourcing our MIS.
We also retained outside counsel, since this was our first time to outsource our IT. We contracted with two consulting firms that employ experts on outsourcing. These outside experts didn't determine our objectives or set the time tables. They simply pointed out the pitfalls of outsourcing. Also, because putting down in writing exactly what we're all agreeing to can be very difficult, we retained two legal firms that are familiar with constructing outsourcing contracts.
And then we arranged to meet with an executive at Eastman Kodak and to use Kodak as a benchmark for Bethlehem Steel's outsourcing project. Kodak has already outsourced its MIS function, and we knew we could learn from its experience.
As for our pick of outsourcer, we decided early on that we didn't want to fragment our outsourcing in a way that would suboptimize the outcome. We wanted to name one partner and to charge that partner with the responsibility of doing what's needed where it's needed and to avoid having competing internal forces vying for space on the mainframe and on the client/server. For us, that meant the outsourcer should have control over the data centers, all of our application development, our personal computers, our process computers and all of our networking activities, both local area and wide area.
Finally, in analyzing the financial aspects of the outsourcing proposals, our benchmark was the amount we figured we would have to spend to reduce costs and at the same time maintain our technological base on our own. We came up with about $10 million a year for a data center, capital expenditures, overtime, consolidation and so forth. We hired an outside consultant to study this benchmark and verify that our projections TABULAR DATA OMITTED of our future costs -- and any future reductions in the number of people we would need -- were reasonable. Thus, that number became our reference mark, and both of the top two bids were compared against it.
THE HOME STRETCH
By mid-November of 1992, we had selected the outsourcer that best fit our needs, Electronic Data Systems (EDS). At that point, we just needed to work out the implementation details: what would happen to our MIS employees, how we would migrate the physical assets and what we would do about application development.
First, we had to transfer approximately 450 employees from Bethlehem Steel to the outsourcer. Three hundred employees were permanent transfers; 150 were transitional transfers. The transitional transfers would have employment with EDS at least through the migration of our data centers to EDS facilities, after which EDS would make a good-faith effort to find other permanent positions within EDS for these employees. Bethlehem Steel retained 20 technical people to monitor the outsourcing arrangement from our vantage point, and we'll be phasing out 220 contractors. We also are retaining six staff people and eight communications people in our corporate offices. And we continue to employ three or four people on site at each plant, each of whom has one function: to make this partnership work, not to second-guess the outsourcer.
Within the small group of people we retained are one financial person, to interact with the outsourcer on the finances of this very complex project; one application development person; one client/server person; and one process control person. These Bethlehem Steel employees have counterparts employed by the outsourcer at very high levels. We retained no employees at the lower levels.
As a part of the employee transition process, we asked our human resources staff to wade through, line by line, both our benefits plan and the outsourcer's plan to decide if the benefits were comparable. Fortunately, they were. So we were satisfied that our employees would have few complaints about the benefits they would receive when they were transferred to the outsourcer's personnel roster. Had the benefits not been comparable, we may have faced a problem.
But we still needed to address such issues as how to transfer savings plans and how to handle retirement eligibility. Both of these topics are complex, and fortunately we were able to resolve all of the issues that arose.
A final note on employee transition: One of the common misconceptions about outsourcing the MIS function is that the internal MIS group isn't doing the job. That certainly wasn't the case at Bethlehem Steel. In fact, the MIS professionals were doing a very good job, given the company's constraints: a lack of capital, an inability to attract new people and a reduction in head count. But that's one way you can convince your MIS people, who may not welcome the idea of outsourcing, that the change is a good one. Outsourcing will help them do a better job by giving them more to do their job with, plus it's an opportunity for them to develop professionally since they'll be moving to an IT company.
When we finished addressing the employee transition process, we moved to how we would migrate our two computing centers to our outsourcer's data center. Under our arrangement, EDS assumes responsibility for managing, maintaining and, most important, refreshing Bethlehem Steel's MIS physical assets. Frequently, companies will sell their assets to their outsourcers to raise a little cash. In our case, that made no business sense. Instead, we leased them to EDS for one dollar. As those assets become surplus or outdated, EDS will replace them, and the ownership for those assets will pass to the outsourcer.
Finally, our outsourcer is now responsible for infusing technology into our operations. However, while EDS will maintain all of our existing systems, we won't rely exclusively on the company for new application development, although EDS will be our preferred provider.
Who sets the priorities for systems projects each year under this new arrangement? Bethlehem Steel does. Of course, our outsourcer comes to us with ideas on how we may be able to work differently -- and sometimes those suggestions will change the way we prioritize our work -- but we still are responsible for setting initiatives and justifying them.
One significant difference in this area is we now call the president of each business unit at Bethlehem Steel and ask how much he or she wishes to spend on IT. If, for instance, it makes good business sense to plow in an extra $3 million this year on a particular project, that decision now is entirely in the hands of each business unit president. In the past, we were strictly a corporate-controlled entity. IT budgets would be submitted to the corporate office, where they would be arbitrarily slashed and then sent back to the business units. With this method, some good projects would survive while, unfortunately, others would die. We no longer have that problem.
By December, we were in the home stretch. We had established our outsourcing relationship and were satisfied that we had an excellent opportunity to make substantial progress in our business. We completed the contract on December 20, 1992, and announced the outsourcing agreement to the entire company on December 21, 1992. Although we planned to begin outsourcing on February 1, 1993, we were able to start on January 1.
Because we realize 10 years is a long time, we made sure that our contract with the outsourcer includes annual reviews, at which time we can evaluate the progress of the project. However, as I explained earlier, although the contract has escape provisions at various stages if we need them, I cautioned the executives at Bethlehem: Once you commit to this project, you're in. Don't kid yourself into thinking that if we encounter problems we'll bring the MIS function back into the company, because that's very, very unlikely. Fortunately, if one outsourcing partner can't do the job, other outsourcers can.
The key is to understand all of these aspects of the outsourcing relationship and to make provisions for them up front. I'm convinced that because we took a long time to walk our way through every possible contingency, working out the details ahead of time, our partnering with an outsourcer has been a non-event. No one knows the future, that's for sure, but Bethlehem Steel is beginning this outsourcing project very confident that the partnership will work.
Dr. Bargeron is vice president of technology and quality assurance at Bethlehem Steel Corporation in Bethlehem, Pennsylvania.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Information Management; Bethlehem Steel Corp.'s outsourcing of its information management function|
|Author:||Bargeron, Walter N.|
|Date:||May 1, 1993|
|Previous Article:||The trick to managing cash? Be creative.|
|Next Article:||Weaving your benefits into your profits.|