dramatically changing the role of the financial
professional from that of information recorder to business
strategist—making the financial manager much more critical to
the success of an enterprise.
TO KEEP PACE WITH these changes, the financial professional is expected to provide accurate and timely financial information that can be accessed and analyzed quickly and easily. While digital technology may make it easier to collect information and move it from one place to another, it also has led to an incredible proliferation of data. Filtering, sorting, compiling, analyzing and disseminating financial data in ways that add real value to a corporation has become a daunting challenge.
MICROSOFT CORP.— with 54 financial groups charged with providing financial support to more than 85 global subsidiary operations—has struggled with these challenges. Its answer is the financial “digital nervous system,” an intranet-based environment that links all of the company’s financial groups into a single, coherent system and provides its employees with real-time access to information and financial reports through the Internet.
|FIVE YEARS AGO , it took Microsoft two weeks to
close the books. Now it takes four days. The company used to
print and distribute 350,000 hard-copy management reports each
year. Today, none. Through FinWeb, a network of intranet
sites, its employees can submit travel-expense reports and be
reimbursed, purchase goods and services and transfer capital
assets—all from their desktops. They’ve reduced paperwork,
transaction time and publishing and distribution costs. |
IT’S POSSIBLE for any of its employees who need financial information for decision making to access detailed reports that are updated daily. The financial system lets people drill down through layers of information to get answers—quickly, easily and without computer programming skills. None of the technology used to achieve the objectives is beyond the reach of any organization—large or small.
AS A RESULT, the company is able to achieve something finance organizations strive for: the ability to add more value at the strategic end of the business and spend less time processing transactions.
|SCOTT M. BOGGS, CPA, is Microsoft’s corporate controller. Prior to joining Microsoft, he spent eight years with Deloitte, Haskins & Sells as a manager in the emerging business services group.|
T echnology is providing the tools that are revolutionizing the role of financial professionals from that of information recorders and processors to business strategists—making them much more critical to the success of an enterprise.
The changes started about a decade ago, when the personal computer began to supplant the mainframe computer. The desktop machine was suddenly available at the right price and with enough computational power to make it an essential business tool—streamlining manual processes, managing data and exchanging knowledge. And when the PC was linked to an enterprise network, the evolution became a revolution, totally transforming the business environment and the role of the financial professional. Add the incredible growth of the Internet and we have the communication medium for a true global economy in which data can be transmitted around the world almost instantly. Suddenly a company's value is determined not only by its tangible assets but also by its ability to collect and use information.
Business continues to face other transformations triggered by widespread computerization. The number of distribution channels is growing, forcing most companies to increase their mix of products and services—and making the task of keeping track of revenues, expenses and other financial information far more complicated. At the same time, many companies are finding that to stay competitive they must update their fundamental business model to adapt to the ever-changing marketplace.
The financial professional is expected to keep pace with these changes and provide accurate and timely financial information that can be accessed and analyzed quickly and easily. While digital technology may make it easier to collect information and move it from one place to another, it also has led to an incredible proliferation of data. Filtering, sorting, compiling, analyzing and disseminating financial data in ways that add real value to a corporation have become daunting tasks.
At Microsoft, where 54 financial groups are charged with providing financial support to more than 85 global subsidiary operations, we have struggled with these challenges. Our answer is something we call the financial "digital nervous system," an intranet-based environment that links all of the company's financial groups into a single, coherent system and provides our employees with real-time access to information and financial reports through the Internet.
Exhibit 1: Data at the Desktop
Microsoft’s FinWeb, an intranet system that allows employees to get detailed financial data about the company, provides data at the desktop. It also lets them submit travel-expense reports and purchase goods and services. Result: reduced paperwork, transaction time and publishing and distribution costs.
This system—called FinWeb, short for Financial Information Network—was launched in 1995 and is part of a companywide effort to build a larger digital nervous system at Microsoft (see exhibit 1). Through FinWeb's network of intranet sites, our employees can submit travel-expense reports and be reimbursed, purchase goods and services and even transfer capital assets—all from their desktops. We've reduced paperwork, transaction time and publishing and distribution costs.
Most important, we've made it possible for any employee who needs financial information for decision making to access detailed reports that are updated daily. Our financial system lets people drill down through layers of information to get answers quickly—without computer programming skills.
As a result, we've been able to achieve something finance organizations strive for: the ability to add more value at the strategic end of the business and spend less time processing transactions.
STEPS TO DEVELOPING A DIGITAL INFORMATION SYSTEM
financial digital nervous system begins with a basic computing
environment that includes a network and intranet infrastructure.
Users must have common desktop applications. Once these pieces
are in place, the system can be developed following these steps:
n Create a data warehouse by extracting and summarizing transaction data from the general ledger system to an ODBC-compliant database. Most common general ledger applications provide a variety of ways to export data into other applications. Depending on the nature of the extraction, data can be summarized as extracted or processed within the data warehouse.
n Implement pivot table functionality to create standard management reports. Data fields can be organized into rows, columns and page fields, and reports can be formatted to match existing reports. Users with some advanced programming knowledge can use additional tools within their spreadsheets to create custom queries against the data warehouse.
n Create a Web page that shows an index of available standard reports. Icons can be created to open spreadsheet files from the network. Additional Web pages can be created to describe major accounting policies, procedures, assumptions and other relevant information.
n Automate routine transaction processes by creating forms to perform tasks such as the creation of master data and updates, journal entries and invoice processing. Most general ledger applications provide a way to import data from external sources. Once created, these forms can be saved in HTML format as Web pages and users can access them through the intranet. Users with some advanced knowledge of databases can program additional front-end validation.
n Eliminate reports and manual processes that duplicate information available on the financial digital nervous system.
The system we have in place now is a far cry from the one that existed when I joined Microsoft as a finance manager in 1993. We had no intranet, no central, direct access point for electronic distribution of financial information. More than 30 separate general ledgers were scattered around the world and were consolidated on a Digital Equipment VAX mainframe. During the month-end close, I'd get a huge stack of reports every day and spend the week typing numbers into spreadsheets. We spent an enormous amount of time and money copying and collating reports and then shipping them around the world by mail, courier and fax.
The key to solving this problem was to create a data warehouse using an extract of the general ledger in a Microsoft Access database. It was relatively easy to link that data to Microsoft Excel to create standard management reports. As the underlying data are updated, the standard reports are automatically refreshed to reflect the most current information. These electronic files are stored on network file servers, where they can be accessed by anyone with the right password.
It took us about four months to build the system. When it was ready, I e-mailed everyone where to find the electronic reports and informed them that the days of printed management reports were over. Within five minutes of sending the e-mail a senior executive stormed into my office, saying the information was totally worthless. The problem stemmed from the cryptic eight-character spreadsheet names: You might open a file named eu_ww.xls, expecting to read a customer unit profit and loss spreadsheet but get a European unit spreadsheet instead. There was no way to know until you looked.
Around that time we were just starting to learn about the Internet, and we realized that we could create an Excel sheet that essentially looked like a Web page, with a list of profit and loss sheets designated by simple, clear names and pictures. Beneath each name was a macro that pointed to a particular spreadsheet on the network. When we sent that Excel sheet out, we were basically launching the first intercompany intranet application at Microsoft.
While that was a big step forward, we still had a long way to go. For example, we still had more than 30 separate legacy databases (those original general ledgers) and a system riddled with multiple charts of accounts and other inconsistent data taxonomies (rules for classifying and naming information). Although we had simplified the distribution process, it was still impossible to access the origin of the information. So in 1995 we made the commitment to purchase SAP R/3 accounting software as our core system. It took about seven months to implement the financial modules. This system allowed us to have a single chart of accounts and consistent views of information across the organization. It also has significantly decreased our support costs by eliminating multiple accounting systems.
From there we developed a number of Web-based applications that automated routine tasks—such as administrative procurements, employee-expense reimbursements, asset transfers and journal-entry postings—and then fed all the transactions directly into the SAP R/3 system. Automating these processes has allowed us to reduce the cost of some transactions by as much as 90%.
The system is set up so that virtually every transaction flows automatically from a Web site on the Microsoft intranet into the SAP system and is extracted on a daily basis into several SQL Server-based data warehouses, which then produce standard reports. As a result, authorized users can access on their desktop computers standardized consolidated revenue reports summarizing sales for broad product and geographic categories and then drill down through the databases virtually to the level of part numbers. More important, it allows users to customize reports and to answer questions on almost any financial issue.
ODBC is the acronym for open database
connectivity. It's a database programming interface from
Microsoft that provides a common language for Windows
applications to access databases on a network. |
HTML stands for hypertext markup language. It's the standard document format used on the World Wide Web. A subset of SGML (standard generalized markup language), HTML defines the page layout, which includes fonts and graphic elements as well as hypertext links to other documents on the Web. Each hypertext link contains the URL, or address, of a Web page that can reside on the very same server or on any server worldwide, hence the "Worldwide" Web.
SQL stands for structured query language (SQL is pronounced sequel ), a language used to interrogate and process data in a relational database. Originally developed by IBM for its mainframe computers, all database systems designed for client/ server environments support SQL. SQL commands can be used to interactively work with a database or can be embedded within a programming language to interface with a database. Programming extensions to SQL have turned it into a full-blown database programming language.
OLAP is the acronym for online analytical processing. OLAP software allows users to quickly analyze information that has been summarized into multidimensional views and hierarchies. For example, OLAP tools are used to perform trend analysis on sales and financial information and to drill down into masses of sales statistics in order to isolate the products that are the most volatile.
DRILLING FOR ANSWERS
Here's how I use FinWeb: Each week, I receive an e-mail message that contains a link to a standard Excel-format report showing actual and budgeted revenue figures by geographic area and product for the month, quarter and year. One report is dated December 25 and covers the first three weeks of the month (see exhibit 2). Because it's nearing the end of the month and it's time to start thinking about the close, it's an appropriate time to survey our financial condition.
With such up-to-date information, I'm able to spot trends—often with enough time to act on them. I notice we're slightly behind budget for the month, but I'm not worried because we have one week left in the month. The original equipment manufacturer (OEM or PC manufacturer) numbers also are behind budget, but I know that a lot of our OEM revenue is billed during the last week of the month, so I'm not worried about that either. And Europe (which is designated as EMEA) is doing great—already ahead of budget.
Further down, I can look at revenue by products. I notice we are over plan for Microsoft Office. Scanning further down I see that the Windows NT operating system is slightly behind plan, and the Windows 98 operating system is lagging. The fact that games are already well ahead of projection indicates that this is turning into a good Christmas season. (While I have called up this spreadsheet using the link embedded in an e-mail message, I also could have navigated easily to exactly the same spreadsheet from the FinWeb home page.)
Next I begin to drill down to see the source of the numbers. The system uses Excel's PivotTable dynamic views, so I can reconfigure the spreadsheet to show details I'm interested in. (For more on PivotTables, see "Add Perspective to Spreadsheets," JofA, Dec.98, page 91.) PivotTable views are an important part of FinWeb, providing an easy-to-use link between the spreadsheet on my desktop and all of the information in the data warehouses. A PivotTable lets me customize a spreadsheet by moving rows and columns and defining the data in those fields. For FinWeb, we've created formatted PivotTable views with preselected fields that users can choose from.
Using the PivotTable feature, I select Europe, in U.S. dollars, and "All" for most of the other fields (see exhibit 3). I see that the Office Standard edition is ahead of plan, while the Office Professional version is exceeding budget by a little less. With a week to go, Microsoft Office sales in Europe are significantly ahead of projection.
I continue to reconfigure the data to find out why Europe is doing so well. For example, a new PivotTable, which includes revenue totals by country, shows that revenue in Denmark is almost double our plan (see exhibit 4). If I slice and dice the data in yet another way, I see that much of that comes from sales of Office.
For still more in-depth analysis, we use an internally developed application called MS Reports. This tool allows me to create custom queries against the underlying data in the warehouse by dragging and dropping fields into a standardized template within Excel. The MS Reports interface shows in plain English the available fields (see exhibit 5). Once the template is populated by the fields I'm interested in, MS Reports automatically resolves it into SQL query language, executes the query against the data and formats the results into a standard Excel report.
So if I want to understand why Office is doing so well in Denmark, I can turn to MS Reports and figure which customer segments are responsible. After setting up a query that looks at Office sales by licensing program, I see that the major contributor was the licensing of Office to corporate customers (see exhibit 6).
It took just a couple of minutes to set up the query and less than a minute to get the results. With these tools, I am not only able to access up-to-date revenue figures quickly and easily, but I also can set up a sophisticated analysis to help me explore the data and understand the numbers. And I can do this without having to understand the programming language used to query the SQL database.
One reason all this runs so seamlessly is because we have worked to make sure that all data that go into the system are created equally. We want to make sure, for example, that when we compare Denmark's revenue figures for Office with the figures from Japan, the numbers mean the same thing. The same must be true for travel and entertainment expenses and every other transaction recorded in the SAP database. Every number one business unit enters must be comparable to corresponding numbers every other business unit enters.
The key to achieving this kind of companywide uniformity is a site called MAP, short for Microsoft accounting policies (see exhibit 7). MAP contains each of the 500 or so accounting policies that we follow, and it helps us ensure that data conform to common rules and taxonomies across the entire company.
Because it is so easy to access and update, and because the information is always current, MAP is an ideal place for posting policies and procedures of all kinds.
FinWeb is more than just a portal for accessing and analyzing corporate financial data. It's the entry point for virtually every transaction. FinWeb's network of Web sites includes MS Expense, which allows employees to submit expense reports and receive prompt reimbursement; MS Invoice, used by Microsoft vendors for submitting electronic invoices; and MS Market, our internal procurement tool.
Another FinWeb feature is Headtrax, which shows the employment hierarchy at Microsoft, making it possible for managers to see which positions are open and which are filled in their group. It lets them make changes within their organizations. With Headtrax, for example, I can move one of my employees to another group or give a raise to someone who directly reports to me. Assuming I've followed all standard accounting and administrative policies, I can take this action from my computer in seconds, without generating any paperwork. (Of course, you can only access this confidential information if you have the proper approvals and clearances.) And, like all FinWeb transactions, any change I make immediately enters the SAP database and is extracted within 24 hours to a data warehouse, where it becomes part of the information pool available to anyone in the company who needs it.
What are the most critical prerequisites for putting such a system in place? There are about a half-dozen, and they fall into two categories: technical and cultural.
To most financial professionals, the technical challenges may sound daunting, but they are fairly straightforward. To build a financial digital nervous system, a company must have a robust computing infrastructure in which all computers share a common operating system and are linked by a reliable network.
Most important to achieving such a system is the cultural context. In many organizations, while the information technology (IT) people understand technology, they don't understand business. Likewise, financial people understand business but may not understand technology. To create a system comparable to ours requires teamwork between the IT and finance people.
Another essential cultural element is executive commitment. There is always a certain amount of discomfort when old systems are discarded and new ones are put in place. To implement the magnitude of change required to build a financial digital nervous system requires a clear mandate and support from executive management. An organization with a CEO who doesn't use a PC probably shouldn't embark on this path.
At the same time, a certain amount of patience is required. Building a financial digital nervous system has been an incremental process: eliminating paper reports, adding a Web interface, implementing financial software. That process must start with a vision: information at your fingertips. Along the way, our idea of what is possible has expanded. Originally, we were pleased to reduce our closing time from two weeks to four days. Now we are challenging ourselves to achieve a continuous close, where information is accurate and current every day of the month.
As we continue on this path, I see many other opportunities to increase the value of financial information. One project is a new data warehouse that will allow us to consolidate disparate types of data and make it possible for users to create the equivalent of a balanced scorecard. Another project is incorporating new OLAP services included in Microsoft SQL Server 7.0 to significantly increase the performance of our data warehouses. Finally, we are implementing middleware—software that will allow us to create Web-based applications in which users can construct queries without ever knowing which specific databases contain the data. Ultimately, we hope to create an authentic enterprise information portal, where any company can have its information truly at its fingertips . n
Exhibit 2: Financial Data in a Flash
Up-to-the-minute financial information comes to Controller Scott Boggs’ desktop computer regularly. In this flash report, he receives Microsoft’s companywide monthly and quarterly results. He can drill down for more detailed information. The data in this exhibit are for illustration purposes only.
Exhibit 3: Putting Data in Perspective
Using Excel’s PivotTable feature, Boggs can reconfigure the data so he can better understand how sales of different products affect the company’s overall revenue.
Exhibit 4: Slicing and Dicing the Numbers
Data in the PivotTable spreadsheet is sliced and diced further to find out what products are pushing up sales.
Exhibit 5: Creating Custom Queries…
For still more in-depth analysis, Boggs uses an internally developed application called MS Reports, which lets him create custom queries against the underlying data by dragging and dropping fields into a standardized template within Excel.
Exhibit 6: …and Getting the Answers
MS Reports’ interface shows in plain English the available data fields. Once the fields Boggs wants to examine populate the template, MS Reports automatically resolves it into SQL query language, executes the query against the data and then formats the results into a standard Excel spreadsheet report.
Exhibit 7: Accounting Policies Database
The key to achieving companywide uniformity with the financial data is an intranet site called MAP. It contains each of the 500 or so accounting policies Microsoft follows, helping ensure that data conform to common rules.