When Swiss Bank Corp. wanted to boost foreign exchange trading in New York, it faced a problem. Traders wanted to cut costs, but the back-office staff simply did not know how the operation stacked up against those of competitors.

Swiss Bank scouted around for reliable information to gauge what it cost its most efficient competitors to process a trade but came up empty-handed. So the bank approached the biggest foreign exchange trading banks with an offer: Show us how your operation works; we'll all share the information; and Swiss Bank will pick up the tab for the research.

More than a dozen money-center and international banks accepted.

"The study helped Swiss Bank tremendously," said Juliet Kontaxis, a principal at Benchmarking Technologies International, a Ridgewood, N.J., consulting firm that was started by Swiss Bank, then spun off. "They were really able to improve their productivity and their costs."

Bankers and consultants said the kind of study initiated by Swiss Bank, known as benchmarking, is a valuable tool to evaluate the efficiency of lines of business. Looking at the productivity, systems, and operations of competitors can be an important first step in shaping business and technology strategies.

But experts caution that the term benchmarking, like the now-ubiquitous word "reengineering," has become a buzzword that can mean many different things.

Not all benchmark studies are created equal, they say, and the pitfalls of the process are not always understood. Moreover, despite benchmarking's popularity, experts say banks have trailed other industries in using the approach to improve operations.

"To do it right involves a lot more money on the part of consultants than banks would have ever guessed going in," said Lawrence A. Willis, executive vice president at First Manhattan Consulting Group. "And so they tend to opt for the cheaper approach to doing it and end up with apples-to- oranges-type comparisons that are really not particularly useful."

Benchmarking studies tend to fall into two broad categories.

The more comprehensive approach, like the one taken by Swiss Bank, involves gathering specific information about how comparable tasks are done at different banks. Consultants meet with bankers, talk about strategies, and collect both quantitative and qualitative data.

Areas examined included strategy, organization, workflow, technology, human resources management, and management controls systems.

This approach, sometimes called "best practices analysis," is more far- reaching than the second category of benchmarking study, straightforward comparisons of productivity numbers. In consultant-speak, the latter measurements are known as metrics.

"How many FX trades per head do you process? How much does it cost to process it? Those are two that they tend to like," said Ms. Kontaxis of Benchmarking Technologies.

Another popular metric is customer service. That's measured, for example, by how long customers must wait before their calls are answered.

"Most of what you see out there are metrics benchmarking studies. They are just raw numbers, basically," said Ms. Kontaxis. "They are not very helpful in terms of giving organizations any information about why their numbers are different from the next guy's. It's just kind of like a report card."

While metrics data can be informative, it's essential that they be reported consistently, consultants say.

"Unless the work is getting done in a very customized way for a particular client, you can get apples-to-oranges-type distortions in the data and just go off totally on the wrong track," said Mr. Willis of First Manhattan.

One risk is the temptation to make hasty decisions to cut costs when a bank finds it fares poorly against competitors. A bank that outperforms the competition on productivity per employee, however, may offer a poorer quality of customer service.

Through its best practices study, begun in 1992,?? Swiss Bank found that one competitor stood out.

That bank was using so-called straight-through processing for foreign exchange trading, which eliminated both data entry employees and paper.

Since then, many banks have introduced such automation, which has greatly improved productivity. Swiss Bank also consolidated two back-office sites into one.

To track improvements, Swiss Bank, along with study participants including Royal Bank of Canada, Citicorp, and BankAmerica Corp., has participated in two metrics studies based on follow-up questionnaires.

For both flavors of study, consultants say it's best to involve at least a dozen banks in order to get a cross-section of industry practices and guarantee participants' confidentiality.

"We certainly need metrics information to really confirm whether the best practice that we're looking at has really led to some sort of performance improvement," said Ms. Kontaxis.

Benchmarking Technologies is now visiting many of the original participants, and several additional companies, for a follow-up best practices study.

On another front, Seafirst Bank, the Seattle-based unit of BankAmerica, was one of 18 banks and finance companies participating in a benchmark study of indirect auto lending. Bill Lauman, senior vice president and manager of dealer banking, said the division has been improving processes continually but recognized that further efficiency gains were essential to its survival.

"If you are not efficient and produce the product at a low cost," he said, "in the next five or 10 years, I don't think you'll be around."

The study was done by Benchmark 2000, a Marietta, Ga.-based consulting firm.

"We get these banks together and, with the same team, go spend a week at each institution," said Randall Foster, benchmark product manager. "You have to do that because it's the only way you can ensure you've got comparable data."

Mr. Foster looks at five key areas: technology, staffing levels, workflow, management quality, and relevant practices specific to each bank.

Differences in bank performance can be enormous.

Seafirst's Mr. Lauman said credit officers at the best bank in the group made 90 loan decisions per day - 10 times the production at the lowest- ranking bank.

"We didn't know what the marketplace was doing," said Mr. Lauman. "This is the first time I've been able to compare, rationally and credibly, productivity numbers. Now I have a realistic target. I can point to my people and say, 'There are people out there reaching these levels.'"

While Seafirst ranked in the top quartile in most functions, Mr. Lauman said, the study identified areas for improvement. "We hadn't staffed as well as we could. Learning that was a real help," he said. "We felt this was a very cost-effective way to get what we needed."

But Mr. Lauman also noted that the benchmark study did not measure everything. For example, it didn't focus on customer service, so he could not tell whether the bank with the most productive credit officers also had the slowest loan turnaround time.

"You can't look at any type of benchmarking exercise as a stand-alone," he said. "It's part of a whole productivity improvement cycle. You have to use it as a learning tool."

And First Manhattan's Mr. Willis cautioned that banks should not base cost-cutting exercises on benchmark data alone.

"You must be able to relate those findings to both the business strategy and the technology positioning of the institution," he said, "in order to really come up with ... recommendations" that could be acted on.

First Manhattan does 30 to 40 studies a year for clients. But its brand of benchmarking compares a bank's operations to a data base of information collected during its strategic consulting work. "We do considerably fewer engagements where the sole objective of the review is to just do benchmarking," said Mr. Willis.

Charles Callen, senior vice president and managing director of CSC Index in New York, cited another potential benchmarking pitfall: that companies in an industry will simply be in a "mad rush to parity."

"We don't want to just look at our peers; we want to look at what best practice there is outside of our industry so that we can get the advantage of some original thinking," said Mr. Callen. "Otherwise, we are just doing what the rest of the industry does."

Some benchmarking efforts have sought participants beyond the banking industry. The study of indirect auto lending, for example, included the consumer auto finance units of General Motors, Nissan, BMW, and Volkswagen.

But bankers and consultants agree that, despite the progress made in improving banks' operating efficiency, much remains to be done. "A lot of the innovation has come from nonbank banks," noted Mr. Callen.

Furthermore, said Ms. Kontaxis of Benchmarking Technologies, "Most of the benchmarking that's been done has not been in financial services. Financial services has been a laggard. Banks are just not on the forefront."

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.