Details, Details

Transaction data doesn't usually play well with others. Here's how some companies are getting around the problem.
Marie LeoneMay 7, 2002

Managing data can be a real pain (your Brent Spiner joke here).

In fact, Alan Yong, a veteran technology analyst at Aberdeen Group in Boston, says he was shocked to discover that users still list data integration as one of the biggest hurdles in implementing budgeting and planning (B&P) software. Even with recent advances in financial portals and application program interfaces, CFOs still report sizable headaches in getting B&P software to access information stored in far-flung computer systems.

Compounding the pain: Many companies are moving away from once-a-year budgeting, choosing instead to go with hybrid annual-rolling forecasts. Those applications, fueled by real-time updates and analysis, don’t offer a heck of lot of insight when parsing data from 1963. “Even top-flight B&P solutions fall short if the legacy system it’s pulling data from is not running in real time,” notes Yong. “The data flow remains disjointed.”

Drive Business Strategy and Growth

Drive Business Strategy and Growth

Learn how NetSuite Financial Management allows you to quickly and easily model what-if scenarios and generate reports.

Very Retro

Typically, that disjointing starts when companies attempt to bring together scads of spreadsheets scattered across scores of departmental applications. As e.Intelligence CEO Richard Tanler notes, it’s hard to know whose data to trust when tens of thousands of Excel files flood in from every corner of a company — often in a willy-nilly fashion. Such ad hoc management of data, says Tanler, leads to all sorts of troubles, from security snags and input errors to spreadsheet formats that discourage adjustments after a budget is in place.

If this all sounds discouraging, buck up. While data may be the root of the budgeting and planning problem, it’s also the solution. Experts say data warehouses — those massive storage devices companies starting buying in the mid to late ’90s — can actually bring a little sanity to the budgeting and planning process. “Companies never thought that the biggest aid to planning would be the data warehouse,” declares Tanler.

Indeed, finance managers have mostly used data warehouses to see what has happened at a company — not what might happen. In fact, Tanler estimates that 80 percent of the data warehouses in use today lack detailed planning data. “If those [planning] numbers exist, they are probably in a spreadsheet somewhere,” he claims. “But they’re not connected to anyone else’s spreadsheet.”

Take the case of Lands’ End, the Dodgeville, Wisconsin-based retailer. For years, management at the direct merchant relied on a homegrown spreadsheet when making budgets. By all accounts, the proprietary program was not nearly up to the task. Patrick Gilbert, director of financial planning and analysis, says the in-house software created a nightmare for finance staffers trying to consolidate information at the department and account level. “It used to literally take days to make simple changes.” recalls Gilbert.

Last June, management at Lands’ End finally decided to ditch its in-house setup. Within five months, the company rolled out an enterprisewide B&P system. Gilbert says it’s proved to be a real upgrade. For example, using the old software, there was no easy way for Gilbert to enter an employee health insurance upgrade that would trigger an across-the-board price increase. “Now,” Gilbert notes, “I can make that kind of change in a heartbeat.”

The Lands’ End financial planner says he had another reason for investing in a new B&P system: He needed better management of an uneven revenue cycle. “Our business doesn’t fit a fiscal timeframe,” notes Gilbert. The company’s call center operates 24 hours a day. Specialty catalogs — which drive additional revenue — supplement regular catalog mailings. And the gift-giving season is no longer relegated to December.

With the vendor-built B&P application in place, Gilbert now issues monthly budget reforecasts that can include consolidated and departmental reports. His next goal is to move into seasonal forecasting, which mirrors how the company’s merchandising and production units operate. Notes Gilbert: “Lands’ End wants to know what’s around the corner.”

Details, Details

The desire to avoid nasty surprises is not limited to senior level executives. Managers at operating units, divisions, and subsidiaries are also generally keen to produce more accurate forecasts — and thus hit their numbers.

Such is the case with Caterpiller Logistics Services, the Morton, Illinois-based unit of heavy equipment manufacture Caterpillar Inc. According to David Swanson, a business analyst at Cat Logistics, the unit’s business resources department collects 550 separate profit-and-loss statements in one standard template. That template then gets rolled up and sent to the corporate database. But unlike users of some older B&P software, Swanson is able to pull down consolidated data from the parent company. He then uses the application to break down the corporate numbers into lower-level detail to improve unit budgeting and planning.

When the data is downloaded, for instance, Swanson can look at consolidated travel expenses by region, department manager, unit vice president, or any other defined parameter. The system also helps with currency translation. That’s important, since management at Caterpillar Inc. wants all foreign unit sales figures adjusted back to the beginning of the year. Without the adjustment, Swanson says it’s not always easy to tell if an operating unit turned in a solid performance — or merely benefited from changes in forex rates.

Of course, there is such a thing as too much information. Even vendors concede that terabytes of data tend to overwhelm — not enlighten — users. “It’s a double-edged sword,” grants Andrew Ferguson, CEO of vendor SRC Software. “When the focus is to drill down to a very detailed level, there is always a trade-off.” To avoid overloading employees, Ferguson advises clients to figure out exactly what value is derived from the data being crunched. After that, he says users must strike a balance between the need for detailed analysis and the need to actually generate a return on B&P investments. “In this case,” Ferguson notes, “it’s a fallacy to think more is always better.”

Still, it’s not surprising some finance managers stuff their B&P programs with as much data as possible. All things being equal, detailed entries do yield more accurate forecasting models than less specific entries. And as Ferguson points out, finance managers want to feel confident about their numbers when they’re running their budgets by cost-conscious board members.

Aberdeen’s Yong makes another case for keeping more detailed data. “Drilling down to the granular level,” he notes, “allows a company to slice, dice, and customize the information for specific users and set business parameters around the data.”

Indeed, future incarnations of B&P software will probably act a lot like business intelligence software. Howard Lamb, vice president for information technology at Celerant Consulting, says his company is designing several client prototype systems that turn out business intelligence, not just reams of transactional data. The idea, says Lamb, is to design customized, Web-based systems.

While he can’t go into specifics, Lamb offers this scenario. Armed with a sophisticated B&P system, the CFO of a manufacturing company whose operating plants rely on No. 6 fuel oil is immediately notified by E-mail that Venezuela has just stopped oil production. At the same time, the finance chief receives another alert pointing to all relevant real-time internal reports. Armed with that data, the CFO makes an informed business decision.

A Universe of Data

If this seems a long way off, you’d be surprised. Experts say these sorts of hybrid B&P-BI programs are only a few years from delivery.

Whether companies feed those systems the right information remains to be seen. While many businesses are storing more data these days, they’re not always storing the right data. “Garbage in, garbage out is a truism,” says Scott Davis, CEO of software maker Eyeris Inc. “But it’s also a red herring.”

Come again? Davis points out that every one of his corporate clients has managed to muddle through info dilemmas, such as data sets that don’t play well with others. But he says these same clients often ignore the underlying problem: B&P software doesn’t usually cull data from the company’s core operations, marketing and financial systems. If the software did pull records from what Davis calls the “deep, old, rock-solid reliable core source,” the data would not be marred. As it stands, most data still travels through myriad software translators before reaching the B&P system. Much gets lost in the translation.

Ironically, Aberdeen’s Yong blames some of the problems with bad data on the proliferation of information. There is, he says, a major trend within financial departments to import more than just internal financial data into B&P systems. “Once you increase the universe of data to include, say, logistics information, marketing measures, and economic statistics,” Yong says, “everything changes.”

In other words, managing data can be a real pain.

Marie Leone is a senior editor at

A Brick House?

Wesley MacLagan, practice manager for software consultancy Parson Group, offers a rather unusual analogy to explain the corporate approaches to implementing financial planning and analysis software (FP&A). MacLagan says there are three strategies for deploying FP&A software tools (a superset of B&P software): the straw, stick, and brick models.

MacLagan’s Three Little Pigs analogy, while a bit peculiar, is spot on. Many corporate implementers follow the straw model, which ushers a company into the realm of collaborative B&P. But since the software sits on top of a non-ERP-integrated system, the program relies on information gathered from spreadsheets with lots of connecting translation software. Because the structure is loose, it’s difficult to keep it in synch with the changes of the business model, MacLagan says.

The stick model places the software on top of a single-source integrated ERP system. With any luck, the ERP system does represent the user’s business model. That helps employees at a company’s different business units talk the same data language. Getting accounting and sales to calculate margins and profitability the same way is a big leap forward in the world of budgeting and planning.

And the brick house approach? According to MacLagan, that tack relies on a common ERP platform which reflects the company’s business model and exists within well-integrated business processes. Information is therefore available to senior executives, line managers, analysts, and finance staffers. Under this scenario, data moves seamlessly between departments, systems, and users. Moreover, there are few human or system failure points.

Currently, few corporations have embraced the brick house approach. That’s likely to change over the next few years, as corporate finance departments look to involve all of a company’s units in the budgeting and planning process. In addition, champions of the brick house approach say it’s mighty mighty. —M.L.