For Harvard Pilgrim Health Care Inc., it was the best of years, and it was the worst of years. In 1998, Newsweek rated the health maintenance organization the finest in the country, awarding it a string of “A’s” in various categories. Similar accolades came from U.S. News & World Report and a few consultancies, too.
But if Harvard Pilgrim’s service passed with high honors, its business flunked: the HMO posted a $94 million operating loss for fiscal 1998. The deficit was unprecedented, forcing the resignations of CEO Allan I. Greenberg and CFO Thomas J. Brophy last May.
Brookline, Massachusetts- based Harvard Pilgrim, which had actually budgeted for a small surplus, blamed several factors for the loss, including higher-than- expected utilization of medical services, rising drug costs, and losses in its Medicare business–problems endemic to embattled HMOs. But another factor cited is known to companies of all types: “heavy investment in customer service, Y2K, and other systems conversions.”
Translation: Harvard Pilgrim was spending a lot of money exterminating the Millennium Bug. Worse, the HMO, which was formed by the merger of two plans in 1995, still hadn’t integrated its computer systems. In particular, a legacy claims- processing system that previously supported 500,000 subscribers was expected, after the merger, to handle more than 1.2 million–but couldn’t. As a result, “we created a backlog,” says Debra E. Speight, senior vice president and chief information officer.
Speight, who joined Harvard Pilgrim two years after the merger, thought previous stints at Zurich Insurance Group and GE Aerospace had prepared her for anything. But she was wrong. “When I got here, I had no idea!” she exclaims, referring to both the claims-processing problem and the complexity of the HMO’s systems. The backlog, significantly longer than normal, meant medical providers would be paid late, causing them cash-flow problems and creating ill will.
“It’s difficult to understand where you are [in the business], with a backlog,” sums up Speight, who, when asked about its dimensions, replies, “We’re working on getting it down.”
Experienced CIOs like Debra Speight are familiar with the kinds of systems-integration problems that can result from a merger or acquisition, but many non- IT managers aren’t. It’s a dangerous ignorance. “[Information technology] can break a deal if you don’t look at it and make good decisions in due diligence,” warns Eugene M. McQuade, vice chairman and CFO of Fleet Financial Group, a bank that has grown through acquisitions from $6 billion (in assets) in 1983 to $106.2 billion, in June. “It’s typically the backbone of cost saves. If you’re not comfortable with it before the deal, don’t do it.”
These days, of course, no company would acquire another before ascertaining the extent of the target’s year 2000 problems. But millennial health is just one aspect of IT operations that must be vetted in a merger or acquisition.
Square Pegs, Round Holes
Merging acquired systems can be predictable, if not painless. Forrester Research Inc., an IT advisory firm in Cambridge, Massachusetts, classifies mergers and acquisitions into four species, with an integration strategy suitable for each. But no matter how generic the business combination, unforeseen complications may arise in IT.
Potentially fraught with difficulties is the marriage of equals, such as the merger that formed Harvard Pilgrim Health Care. Choices must be made between redundant systems–and personnel. And the issues involved are not just technical but political, too.
Fleet, for example, is currently merging with its crosstown rival, BankBoston Corp.; if the deal closes as planned this fall, it will create the nation’s eighth-largest bank, with roughly $180 billion in assets. At press time, the two banks were in the process of choosing which ATM system to use. The technology favors Fleet’s, since it handles four times the volume of BankBoston’s. But McQuade says the ultimate choice won’t be made by fiat; he, along with Fleet CIO Michael Zucchini, has to build a consensus among the two banks’ decision makers.
Meanwhile, a systems standoff looms in the $14 billion merger of AlliedSignal Inc. and Honeywell Inc., which is expected to close in October. Wall Street has spared no hosannas for this combination, since the two companies’ businesses are complementary in key respects. But their information systems are a square peg and a round hole. AlliedSignal uses enterprise resource planning (ERP) software from SAP, while Honeywell uses Oracle software. AlliedSignal handles most of its IT needs in-house; Honeywell reportedly outsources up to 50 percent of its IT operations.
Citing the merger quiet period, an AlliedSignal spokesperson declined to comment on the company’s strategy for integrating the two businesses. Given their size and complexity, it’s a safe bet neither will switch over its core systems anytime soon. The merging firms may be connected with financial consolidation software, while IT managers look for cost-cutting opportunities in infrastructure management and telecommunications.
Generally, companies with different ERP systems may want to consolidate nonfinancial information through a data warehouse, notes Jim Holincheck, an analyst at IT advisory firm Giga Information Group–though a data warehouse is simply an access mechanism, and the real challenge is to integrate different data sets. The merger partners may also seek leverage in common processes; “purchasing is an obvious one,” says Holincheck.
Same Difference
What about a situation in which merger partners use the same ERP software? It may seem reasonable to assume that two identically named systems controlling identically named processes can be combined without too much fuss. But in reality, customization may have made those systems as different from each other as from another vendor’s software.
Ron Shevlin, a senior analyst at Forrester, observes that levels of system customization may exist within a company–plant by plant, business unit by business unit, country by country. “There’s nothing more misleading than the statement, ‘We’ve standardized on SAP,'” declares Shevlin. “It’s like saying, ‘We’ve standardized on C++.'” There may be some commonality of data structures, he notes, “but the core functionality is customized.”
Still, it helps if merging companies are in the same business. One such company in the early stages of combining two SAP systems is DaimlerChrysler AG, the Stuttgart, Germany, and Auburn Hills, Michigan, automotive behemoth. Formed by the 1998 combination of Daimler-Benz AG and Chrysler Corp., the company will eventually have a single, global SAP system, says Susan J. Unger, senior vice president and CIO. Both companies had “embarked on the voyage of launching SAP” when the merger was announced, she says, but neither had gone so far that their systems could not be put on the same course.
But the company faces gnarly decisions about how to structure financial and business processes, then configure SAP accordingly. “It will take some time to do that,” concedes Unger. In the meantime, the CIO says she’s “pleasantly surprised” at the degree of technological complementarity between the two companies. Both Stuttgart and Auburn Hills have major projects under way, but there’s little overlap between them. For example, an Internet communication system for auto dealers was a lower-priority item for Chrysler, but Mercedes already has this capability, which can be relatively easily extended to the American company.
Of course, an umbrella ERP system isn’t the solution in every merger situation. Take the $7.3 billion acquisition of Browning-Ferris Industries Inc. by Allied Waste Industries Inc., announced last March. Allied Waste, which had previously aborted its own SAP project, subsequently said it would scrap Browning-Ferris’s brand- new, $130 million R/3 system. Both waste-industry firms have grown through multiple acquisitions, but Allied Waste prefers to maintain decentralized management, rather than attempt to standardize hundreds of locations on SAP.
Customer Conflicts
Acquirers can develop a false sense of comfort when a target is in the same business. Systems may be tailored to customers, and a target’s customers may be very different from the acquirer’s.
That lesson was learned two years ago by a large pharmacy- benefits manager (PBM), according to Forrester’s Shevlin. The PBM, with $2 billion in annual revenues, acquired another PBM half its size. The acquirer figured rapid integration and cost reduction would produce immediate benefits, earning Wall Street’s approval.
“Very painfully they learned that both sets of systems had been customized to deal with very different customer bases,” says Shevlin. The larger PBM’s business consisted of custom deals with Fortune 100 companies, while the smaller PBM focused on mid-tier firms and other health care providers. Making modifications to accommodate the new data “was a huge investment,” says Shevlin, “something that wasn’t planned for in the merger financials, and that took a long time.” The company’s stock took a hit and has been slow to recover.
Sometimes, data can’t be converted to the acquirer’s systems, points out Donniel S. Schulman, a partner in Pricewaterhouse-Coopers LLP’s Management Consulting Services practice. “Critical functionality may be missing,” explains Schulman. For example, “I may cut my bills off a pricing matrix that you can’t duplicate.” In the telecom industry, bundled pricing is a huge issue, he notes. Wireline and wireless phone providers are merging, but they have been largely stymied in their efforts to give the customer a single bill for all service.
Clashing Styles
Appraising a merger fit isn’t just a matter of systems and data; technology “style” counts, too. In January 1998, MedPartners Inc. and PhyCor Inc., two physician practice management companies, scuttled their $6.25 billion merger, more than two months after it was announced. Why? “Each company takes a much different approach to business in a number of key areas, including information systems,” said PhyCor chairman and CEO Joseph C. Hutts.
One party to a merger might favor sophisticated technology, seeking strategic advantage, while the other prefers a no-frills approach. This dichotomy characterized the 1998 combination of Citicorp and Travelers Group to form Citigroup, a deal eyed with skepticism by IT experts. Michael Schrage, a research associate at MIT’s Media Lab, mocked conventional wisdom that the merger’s success hinged on the compatibility of Citicorp CEO John Reed and Travelers CEO Sandy Weill.
“They could fall madly in love, become godfathers to each other’s children, and creatively collaborate up the wazoo,” wrote Schrage in Computerworld magazine. “Alas, it wouldn’t make a Pentium’s worth of difference to Citigroup’s aspirations if the company can’t swiftly and cost-effectively integrate its disparate systems.”
Fleet Financial has rejected acquisition prospects whose styles clashed, says McQuade; “their approach was so different that, had we integrated them, we would have taken the value out” of the deal. For example, a bank may provide “excessive technology and customer service” compared with Fleet’s, he says. If Fleet were to ratchet down the technology and service, the bank’s customers might rebel.
DaimlerChrysler’s merger brought together contrasting IT styles, reflecting contrasting business models: Daimler-Benz’s autonomous units, serving diverse businesses, each running its own IT shop; and Chrysler’s centralized approach, emphasizing a lean organization. Susan Unger’s challenge is to impose some uniformity on this far- flung organization, starting with the infrastructure. DaimlerChrysler is also setting up “centers of competencies,” which are establishing companywide best practices for business processes.
Best Practices
Given everything that can go wrong with IT in a business combination, companies should take steps to ensure that most things go right. For starters, they should consider making their CIOs privy to acquisitions in the planning stages, not when the papers are signed. At acquisition-minded companies like Fleet and UnitedHealth Group, technology chiefs enjoy senior-executive status.
Above all, companies should develop an IT merger plan. Typically, integrating an acquisition has four broad phases, says Marc Cecere, a vice president who covers IT management at Giga Information Group.
In the first, premerger phase, when there is a tight lid on communications, “you have to be looking at the ‘outliers’–the benefits packages of IT people, structure of the IT organization, and so on,” says Cecere. As for systems, “you can’t get into [them], because you haven’t signed the deal yet. But you have to ask, What are we doing the deal for? Cost savings? To increase market share? Can those systems and people support that goal?”
The second, planning phase focuses on three things: (1) establishing a merger leadership team, (2) assessing the technology portfolio, and (3) opening up lines of controlled communications from IT to “stakeholders” in a merger, including employees, customers, and suppliers. The latter are frequently neglected, says Cecere, with the result that insecurity and rumors abound.
In the third, implementation phase, “everybody does the same job as before,” says Cecere. “You’re really worried about systems people leaving, and it takes time to put together an organizational design.” As a rule of thumb, he says, 20 percent of IT staffers from the acquired company have to stay–“the high- performing people, those who do network maintenance, operate mainframes, and so on. It’s critical to keep systems running and minimize disruption.”
It’s also vital to establish basic communications between the merger partners–telephone systems, E- mail, intranet connections. “On Day One, it’s really important to tie the communications together seamlessly,” says Unger of DaimlerChrysler, which connected its Internet and intranet sites within six months, and by the first day of the new company.
In the final, postimplementation phase, “you do a postmortem on lessons learned,” says Cecere, adding a sobering coda: “Even if you do everything right, you’re probably going to move up from ‘miserable’ to just ‘wretched.’ But you’ll reduce the misery quotient.”
One Plus One Minus One
The misery hasn’t ended at Harvard Pilgrim Health Care, which followed its dismal 1998 showing with a net loss of $22 million for the first quarter of 1999. New CEO Charles D. Baker Jr. has called in the cavalry–Perot Systems Corp., the Dallas-based business and systems integrator. Every major aspect of the organization, including IT, will be scrutinized.
What postmerger lessons has Harvard Pilgrim learned? Speight emphasizes two. First, scalability issues can’t be overstressed: “One plus one doesn’t [necessarily] equal two. You need computer power.” Second, technology can’t solve business-process issues. “Systems can only reflect the complexity of the business,” says Speight. “If you haven’t rationalized your business processes, your systems are going to struggle.”
At press time, according to a source, Harvard Pilgrim was reevaluating its cure for the claims-processing problem. But Debra Speight won’t be in on the final decision: she resigned at the end of June. New CIO Louis Gutierrez officially took over the IT reins on July 8. S
Edwa rd Teach is a senior editor at CFO.
Four M&A Strategies
How to integrate IT operations depends on the kind of business combination.
- Marriage. Companies are of similar size and play similar roles in the same industry.
Strategy: Merge both the people and the technology. - Acquisition. The acquiring company is at least twice the size of the acquiree–both companies are in the same industry and play similar roles.
Strategy: Merge the people but replace the technology. - Vertical integration. The companies are in the same industry but perform different roles–a manufacturer acquiring one of its distributors, for example.
Strategy: Merge the people and build systems interfaces. - Conglomeration. The companies are in different industries altogether.
Strategy: Maintain separate IT groups and infrastructures.
Source: Forrester Research Inc.
Banking On IT
Fleet’s fast growth depends on techies.
Information technology is crucial to all businesses, but particularly so in such information-intensive industries as financial services, health care, high tech, and telecommunications. All four sectors have witnessed intense M&A activity in the past few years, putting a premium on the ability to integrate acquired IT assets.
As a result, big systems integrators like IBM, EDS, Andersen Consulting, and PricewaterhouseCoopers are much in demand–not just for mergers but also for joint ventures, points out Donniel S. Schulman, a partner in PwC’s Management Consulting Services. (PwC also helps companies selling business units to ensure that the sold units will have autonomous operations, adds Schulman. “It’s typically cut into the deal.”)
What’s more, organizations that regularly make acquisitions now view systems integration as a strategic competency. Companies like First Union Bank, UnitedHealth Group, Tyco International, and Cisco Systems have developed what are essentially in-house SWAT teams for IT. The swifter they can merge IT in an acquisition, the sooner they can reap benefits from the deal.
One fast mover that lives up to its name is Fleet Financial Group, which considers its in-house integration abilities “one of our strategic strengths,” says vice chairman and CFO Eugene M. McQuade. “We devote mind- boggling resources to this capacity.”
Those resources have been put to the test again and again, notably in the $3.7 billion merger with Shawmut National Corp. in 1995, the $3.6 billion merger with NatWest National Corp. in 1996, and now the $16 billion megamerger with BankBoston Corp., scheduled to close in the fall of 1999. “We’ve shied away from small- bank acquisitions in the past five years or so,” McQuade says, noting that integrating a small bank requires “80 percent of the work [to integrate] a large bank,” with the concomitant “huge disproportion in value.”
Core deposit and loan systems always have to be merged, he says, as well as any other system with the potential for customer overlap. But, McQuade adds, “we may not touch” stand-alone operations, such as a private- banking system or a securities-trading unit. Typically, though, Fleet discards acquired applications. When there is a choice to be made, the bank’s priorities are, first, maintaining a seamless presentation to the customer; and second, ensuring optimum cost and efficiency, which includes system capacity.
Data conversion is a critical task, but the hardest job, says McQuade, is maintaining “what we call the plumbing here–the interfaces.” All of Fleet’s systems are connected with either the general ledger, asset/liability, or credit systems. Also, its data warehouse extracts information from hundreds of sources. “Because systems are integrated over a period of time, you end up in many cases doing different rewrites of feeds from core systems,” McQuade says.
Fleet has kept IT turnover relatively low over the past 10 years, helping it grow from obscurity to, soon, the nation’s eighth-largest bank. McQuade offers this salute to Fleet’s techies: “We have made a significant number of M&A bids because we are so confident of our technologists’ ability to deliver on integration plans and cost saves.” — E.T.
Merger Machine
A health giant absorbs IT at ER speed.
Can you name a company with 5 percent greater revenue growth than Microsoft Corp.’s over the past dozen years? If you said UnitedHealth Group, you’re correct. The Minneapolis-based diversified health enterprise’s dizzying ascent from $200 million in revenues to its current $19 billion reflects a frenetic acquisition pace–about 15 in the past few years alone, according to CIO Paul F. LeFort. The acquisitions have ranged from managed health care plans to publishing concerns, from clinical- research organizations to software companies.
The systems of all those companies had to be absorbed, and that’s why UnitedHealth has a team of about 80 IT employees who work full-time on acquisitions or integration. They immediately take over the day-to- day operations of an acquired health plan, working down backlogs as soon as possible. The team follows what LeFort calls a “cookbook approach” to systems conversions. “Our philosophy is to bend operations around the systems when we do consolidations,” he says. “Ninety percent [of any changes] are in operations, not systems.”
UnitedHealth’s highly structured due diligence process focuses in turn on applications and infrastructure (telecommunications and networks, hardware, and middleware). LeFort says the company takes an “absolute approach” to financial and human-resource programs, switching all acquired companies to its PeopleSoft system and Hyperion consolidation software. But with other applications, it will mix and match when feasible, taking pieces of acquired systems and bolting them on other systems.
Claims-processing and enrollment systems are particularly daunting to integrate, says LeFort. Yet, since 1995, UnitedHealth has smoothly migrated millions of customers from a dozen different claims engines to its own two systems. “We moved MetraHealth [United-Health’s biggest acquisition, with 8 million to 9 million members, in 1995] over three years, and nobody knew it was happening,” says LeFort with pride.
Where UnitedHealth reaps the greatest IT cost savings from an acquisition is the infrastructure. “About 50 percent of your costs are on the infrastructure,” notes LeFort. “If you can get phone calls [from an acquisition] at 3 cents instead of 10 cents, you get immediate savings.” The company outsources its telecommunications and data-center operations to AT&T and, primarily, IBM–making it easier to switch acquisitions to a shared infrastructure, and supplementing UnitedHealth’s own IT expertise with what LeFort calls his “strategic partners.”
To would-be acquirers, LeFort offers three pieces of advice. One: “Ask, What are the cost-reduction synergies? If you make an acquisition, can you load what that company does on your infrastructure? ” Two: “Understand the strength and stability of your systems. Will one more straw break their back?” Three: “Make sure you’re buying some of the key talent, what I call the ‘thoughtware’ assets. You really need those.” LeFort knows: he himself came to UnitedHealth via acquisition, as the CIO of MetraHealth. — E.T.