An appetite for growth

www.cio.com –

YOU MIGHT CALL Lloyd DeVaux the wizard of bank acquisitions. In 1998 alone, DeVaux, executive vice president and CIO at Union Planters, oversaw the technical side of 18 different bank acquisitions. In 1999, there were five more acquisitions. In fact, DeVaux has been involved in more than 50 bank acquisitions since he joined the $33 billion Memphis, Tenn.-based bank holding company in 1994. It's enough to make anyone's head spin.

But here's the kicker: These acquisitions didn't happen one at a time, in orderly fashion. Between 1989 and 1999, one of the growth strategies Union Planters pursued was continuous acquisition -- that is, it was in the process of acquiring various smaller banks all the time. Many acquisitions were underway at once, as were multiple systems conversions. And the banks being acquired were scattered across the country. The result? Lloyd DeVaux was one very, very busy man.

Continuous acquisition is a strategy that many companies -- particularly in the banking industry -- have chosen to pursue as a means of growth. "It's a lot easier to grow by acquisition than it is to grow organically by building new banks and trying to win new depositors," says Mark Feldman, partner and managing director of the global M&A consulting division of PricewaterhouseCoopers in San Francisco. But a continuous acquisition strategy is much less certain of success today than it was a few years ago, because the pace of acquisitions has driven up prices for smaller institutions, says Feldman, who is also coauthor of Five Frogs on a Log: A CEO's Field Guide to Accelerating the Transition in Mergers, Acquisitions, and Gut Wrenching Change (HarperCollins Publishers, 1999).

Indeed, Union Planters itself has not made any new acquisitions recently (DeVaux declines to speculate about the company's future plans), and it's not yet clear whether the strategy has paid off for the bank. The 18 acquisitions in 1998 alone increased UP's assets by 78 percent, from $18 billion to $32 billion. But in his letter to shareholders in the company's 1998 annual report, Chairman and CEO Benjamin W. Rawlins noted the company had achieved only half the $100 million savings in operating costs it had projected from those acquisitions. That's largely because in many cases UP had to keep redundant staffs in place for longer than it had anticipated. Expenses during the acquisition period were also higher than anticipated, Rawlins noted, though the company expected to recoup the projected savings in 1999. As a result, profitability did not grow as quickly as expected.

Yet even with the chaos that just a single acquisition can cause, UP has managed to preserve the good graces of its customers. Customer retention has been better than average for a company in the middle of continuous acquisition, says DeVaux.

In any case, the past few years have made Union Planters in general and DeVaux in particular an authority on continuous acquisition. The following are best practices the company has culled for simultaneously managing multiple acquisition projects.

Spot the Bad Deals

UP's continuous acquisition strategy was simple: Buy a smaller bank, install standard IT systems and operate the newly merged entity much more cheaply than it could be as a separate organization. In general, the banks acquired were small compared with UP, and they often outsourced their IT operations. So, for UP, eliminating the acquired bank's outsourcer and bringing IT back in-house promised to save a lot of money. Overall, the idea was to gain the target bank's assets (and customers) while losing much of its expenses. In addition, the acquisitions gave UP an instant customer base in areas of the country in which it had not previously operated.

In order to reap those benefits, though, Union Planters had to pay careful attention to making sure -- in advance -- that a deal was right. As CIO, DeVaux was often part of the UP executive team that went onsite to evaluate target banks in advance of any deal, and his input was considered in the decision whether or not to go through with the proposed acquisition. As such, it was his job to speak up -- before both sides signed on the dotted line -- about any potentially serious systems problems.

For example, for each proposed acquisition DeVaux estimated the value of buying out the target bank's outstanding service contracts, such as PC hardware leases or mainframe application maintenance. If there was a stiff penalty to void one of the target bank's contracts -- say, a $1.5 million price tag to walk away from a mainframe lease -- and the value of the deal as a whole was small -- say $500 million -- that would be a deal killer.

DeVaux had to develop a clear picture of where he could save costs (by eliminating redundant expenses, for example) as well as increase revenues (such as by picking up the target bank's ATM fees). The fact that shareholders in the banking industry expect to see payback on the acquisition within a year only increased the pressure on the up-front numbers. "The payback period used to be a couple of years. Now, the investors are expecting a one-year payback on the acquisition," says DeVaux.

Take the Cookie-Cutter Approach

DeVaux is thankful that as a rule UP acquired smaller banks, obviating any questions about keeping part or all of a target bank's architecture. To maintain economies of scale, UP chose to use the same IT systems in all of its branches, so the logical choice was to replace the target bank's systems (if it had any that weren't outsourced) with its own. As DeVaux puts it, "when the big fish eats the little fish, it doesn't adopt the little fish's systems."

DeVaux's team extended a frame relay WAN link to the target bank, installed Cisco routers and other networking equipment and then installed other systems such as PCs. UP uses Kirchman Corp.'s Dimension 3000 running on an IBM mainframe for its core banking applications. ATMs and bank teller equipment are often an exception to the clean sweep.

This scenario is much easier than a merger of equals. "They can just install their standard systems and go," says PricewaterhouseCoopers' Feldman.

Don't Create a Dedicated Team

Throughout its 10 years on the acquisition track, UP never created a separate team to carry out all the data conversions. This may seem counterintuitive, but both DeVaux and Margaret Aiello, senior vice president of retail branch support and chief project manager in charge of systems conversions for the bank's acquisitions from late 1998 through 1999, are convinced the better approach is to create the conversion team for each project on the fly, staffed by bank employees who have conversion experience.

The idea, says DeVaux, is to have a little extra "bandwidth," or manpower, in every department than is necessary for regular business. That way, during conversion projects, each department can spare some staff time for a project as needed. Once the project is over, the team members return to their ordinary lives, until the next time they're needed.

UP's temporary teams had many advantages over the dedicated team approach, say DeVaux and Aiello. The primary one is that people on the team were current on UP's Kirchman banking system, since they used it daily in their regular jobs and didn't need aany training, DeVaux says. Members of a dedicated team, on the other hand, might not use the system themselves in a production environment and might not know it as well.

That intimacy with UP's systems, in turn, had its own advantages. Most notably, UP's conversion team members knew what was normal and what wasn't, and so could help Aiello (who was involved in every conversion) sort out a real emergency (such as all of the teller platforms going down) from a manageable problem (such as a bank employee unable to access his system).

The temporary-team approach also reduced the potential for burnout, since no one except Aiello was involved in every conversion.

Secure the Help You Need

When UP acquired a bank, a number of things could happen with the target bank's IT staff, depending on the size of the bank and its proximity to one of UP's operations centers, among other factors. Across all the banks UP acquired, some IT staff were retained, some would leave voluntarily, and others who were in redundant positions would be given severance packages. No matter what the particular situation at a given bank, though, UP needed to make sure it had the help it needed for the length of the conversion.

DeVaux and his manager of information services, Morris Proctor, had this aspect of the project down to a science. First, they figured out who they had to have in place until the conversion. That usually boiled down to a matter of title, and it was often one IT supervisor per site. They then made financial arrangements with that person -- in the form of bonuses of varying amounts -- in compensation for staying until a certain date. With the tight IT job market, most people jumped at the chance to "double dip," according to DeVaux -- that is, after their time at UP had expired, to continue collecting their bonus while earning a salary at a new job.

Be Organized to a Fault

Although she has no formal project management training, Aiello has organization in her blood. She would have to, given the number of variables she learned to manage at once. Throughout 1998, Aiello ran five to six projects at a time. With that kind of volume, some things are bound to fall through the cracks. Luckily, Aiello always had a local project manager assigned to each conversion project who took care of the finer-level details, such as managing people's schedules and making travel arrangements for the conversion team. Aiello, meanwhile, focused on the high-level project management tasks.

"I [was] the big-picture person who [made] sure all the projects [were] on target. If we had resource issues, I would run interference to make sure those were dealt with," says Aiello. It's critical that the project manager have enough clout to get extra money or bodies if necessary.

Conversion weekends put Aiello's organizational skills to the test. A couple of days prior to the weekend, if not earlier, contractors would install wiring and telecommunications at the target bank; usually this could be accomplished without any disruption of business. Then, after the close of business on the last day before conversion weekend, the target bank would store all of its data on tape to be flown -- generally via charter flight -- to UP headquarters in Memphis.

Of course, this did not always run smoothly. Aiello remembers a conversion of a bank in Indianapolis that got delayed due to a three-day snowstorm a few weeks before the planned conversion date. "We didn't count on that storm when we made the schedule. We had to back everything up," she says. When you've got numerous projects under way, any delay is especially painful since it will affect everything else.

While the conversion team installed the systems during conversion weekend, another team would converge in a nearby war room or operations center to wait for the Mayday calls to roll in. A hot line specially set up forr the purpose gave everyone involved in the conversion project status updates as each piece of the process began or ended.

Aiello insisted on high-end catered meals for the conversion teams -- breakfast, lunch and dinner. Coffee flowed around the clock and candy bars were always close at hand. "We found if you feed them well, it helps keep them happy," she says.

Don't Be Afraid to Abort at the Last Minute

Along with Aiello and heads of the business unit involved in a given conversion, DeVaux was part of the executive team that continually assessed organizational readiness for the conversion. Meetings were held at least weekly, and a final "yea or nay" decision would be made roughly two weeks prior to the scheduled conversion. "We'd sit down with the management team and see if everyone could sign off on going ahead with the weekend," says DeVaux. "We did a thorough investigation of all pieces of the conversion," such as the status of the system testing and training efforts.

The team would delay a conversion if conditions warranted, such as if the target bank couldn't get the data unloaded and back to Memphis in time. For DeVaux, it was a continual balancing act: weighing the costs of delay against the costs of pressing on with a doomed project. And he has delayed two or three over the years due to serious problems such as corrupt data. Although he caught flak from the other executives, he believes he averted bad customer experiences such as lost funds or wrongfully bounced checks. As long as customers haven't suffered, he figures he can sleep at night.

No conversion will happen before its time, he says. "If you pull the plug on a conversion that's supposed to happen, you'll have egg on your face. But if you pull the trigger on one before it was ready, you're going to have blood on your face."

Insider: How the basic tech behind the Internet works
Join the discussion
Be the first to comment on this article. Our Commenting Policies