7.1 The Impact of Industry, Competitors, and Company Business Design on Business Intelligence Opportunities

The North American Industry Classification System (NAICS) is an industry classification system that groups businesses into industries based on the economic activities in which they are primarily engaged (National Technical Information Service. NAICS, 1998). It is a comprehensive system covering the entire field of economic activities, producing and nonproducing. There are 20 sectors, 96 subsectors, and 1,170 industries in NAICS United States, and companies are grouped within these according to the similarity of the processes used to produce goods or services. For example, food manufacturing is a batch-oriented process manufacturing operation in which various ingredients are blended and converted into finished products. Food manufacturing is very different from other industries such as making fabricated metal products, which is a batch-oriented, discrete manufacturing operation with different parts requiring different operations performed by computer-controlled machine tools.

Because of the differences in production processes across industries and sectors, companies that operate in different industries tend to develop different vocabularies around the core operational processes of the business. For example, although food manufacturers talk of recipes and conversion processes, metal products manufacturers talk of bills of material and machining operations. Different production processes also cause industries to develop different performance measures, such as revenue or cost per ton-mile in the shipping industry or revenue per available room (revpar) in the hotel industry.

In addition to industry differences arising from fundamentally different production processes, industries also vary (Porter, 1980) with respect to the following:

These differences between industries impact what BI information is relevant within a given industry and for a given competitor at a given time, thus raising the question posed earlier: Can we usefully generalize about “how to do BI when treatments for various industries may be so different”? These differences suggest a broader strategic line of inquiry that considers the impact of industry, competition within the industry, and a given company’s business design on the BI opportunities for a given company within a given industry. In our view, looking at industry alone is too static because there are different ways of competing within the same industry. Those methods of competition can change over time as industries themselves evolve and mature and as their customers’ needs evolve.

For example, Michael Porter (1980) advances the view that there are three generic competitive strategies: cost leadership, differentiation, and niche focus. Although written some time ago, his observation still holds true. He points out that “each of the generic strategies involves a fundamentally different route to competitive advantage” and that the “specific actions required to implement each generic strategy vary widely from industry to industry, as do the feasible generic strategies in a particular industry” (Porter, 1985). Porter further asserts that competitive advantage “stems from the many discrete activities a firm performs in designing, producing, marketing, delivering, and supporting its product” (Porter, 1985). By using the value chain construct, Porter then leads us to understand that there are many different combinations and permutations of ways that firms and entire value chains (now called supply chains) compete, pointing out that value chain activities such as logistics, operations, marketing, and procurement, among others, are “the building blocks by which a firm creates a product valuable to its buyers” and that “any activity in the value chain can potentially contribute to differentiation” (Porter, 1985).

Porter’s seminal thinking about strategy and industrial economics has spawned elaborations, variations, and applications of his ideas that further suggest the need to customize BI for a given company competing in a given way within a given industry. Recognizing that industries are dynamic and their futures can be shaped, Gary Hamel and C.K. Prahalad (1994) argue that “companies not only compete within the boundaries of existing industries, they compete to shape the structure create differentiation and competitive advantage, as the company Owens & Minor has done in the medical products distribution industry. It uses a BI system called WISDOM (Web Intelligence Supporting Decisions from Owens & Minor) both to provide decision support information to its managers and employees and to help manage its supply chain.

As industries evolve, competitive positioning and the means of competing also evolve. Richard D’Aveni (1994) points out that “the most important aspect of competition is not current position, but the changes created by the dynamic interaction between rival firms” and that “cycles of change are growing progressively shorter. Although it may be difficult to generalize about how short change cycles are, because cycle duration varies by industry, there is little doubt that what customers value changes over time, and thus the ways companies compete must also change. Adrian Slywotsky (1996) talks in terms of “value migration” and the evolution of “business designs,” where value migration occurs because customers’ most important needs change. To meet those changing needs, companies’ business designs must change. Customers give their business to the companies whose business designs best meets their needs, thus creating value migration from obsolete business designs to more effective business designs.

The concept of a business design is of particular importance for BI purposes. Slywotsky defines the business design as “the totality of how a company selects its customers, defines and differentiates its offerings, defines the tasks it will perform itself and which it will outsource, configures its resources, goes to market, creates utility for customers, and captures profit.” To illustrate, let us consider Wal-Mart, whose business design is summarized in Table 7-1.

For many of the elements of business design, Wal-Mart could have made different choices. For example, it could have chosen to compete on service, as does Nordstrom, or it could have concentrated on consumer electronics, as does Best Buy. For any given company, the choice of business design dictates the performance objectives of core business processes and the elements of business performance that are important to plan, measure, control, and improve. To continue with the Wal-Mart example, its business design dictates that it be concerned with core business processes such as demand planning, forecasting, purchasing, supply chain management, category management, and cost management. That focus drives BI opportunities such as those shown in Figure 7-1.

In other words, the choice of business design suggests areas of analysis for determining the specific BI opportunities that are strategically and operationally relevant for a given company. Given that business designs vary by company and industry, we certainly cannot say a priori how a company in a specific industry should use BI. That being said, we can generalize that for any given company in any given industry, we should systematically evaluate its industry, strategy, and business design as a means of identifying potential BI opportunities. This is the essence of business-centric BI development methods such as the BI Pathway.

Table 7-1
Sample subjects and techniques for business analysis

More broadly, we have tried to establish that industries are dynamic, that the ways that companies compete within industries and industry segments are variable and dynamic, and that business designs are variable and dynamic. All those facts point to the conclusion that the way a given company uses BI ought to be customized to its specific circumstances. That said, industrial economics is industrial economics, and all industries have customers and all companies use inputs and processes to deliver products or services to meet the needs of those customers. Further, the functions within businesses are common, for example, research and development (R&D) or service line development, operations, marketing, sales, finance, human resources, and administration. So although the permutations and combinations of value chain activities are many and although the bases of competition are different within different industries at different stages of industry maturity, it is fair to say that the building blocks with which a given business design is built are common piece parts. Accordingly, what we look at in determining the appropriate BI investments for a given company is the same, even though the optimal BI portfolio for a given company in a given industry will certainly vary. We described an approach for this BI opportunity analysis in Chapter 2, and we elaborated on the subject in Chapter 4.

FIGURE 7-1 The connection of business design, business processes, and business intelligence opportunities.

We talked earlier about companies needing to decide whether they want to be innovative and imaginative in their use of BI or whether they can safely be content as followers, fast or otherwise. For either path, it helps to understand the state of the art, either as a means of stimulating thinking about how to innovate or as a means of understanding the path to follow. Toward that end, the remainder of this chapter will describe important ways that BI can be used to drive profit improvement, starting with a general overview.

7.2 A General Overview of Business Intelligence Opportunities

If we think of BI as business information and business analysis in support of fact-based decisions in the context of business processes that impact profits, it quickly becomes clear that BI is a broad concept. The nature of business information varies along a number of dimensions, including

In addition to the wide range of types of business analyses, a broad range of business analysis techniques may be used alone or in combination with each other to support business analysis. The appropriateness of these techniques vary, based on the subject of analysis, for example, data mining techniques applied to risk analysis in contrast to trend analysis applied to sales performance analysis. Table 7-2 shows a sample list of analytical subjects and techniques.

As the above lists suggest, there are many different possible combinations of types of business information and business analyses. Selecting the relevant business information and analyses is a function of the kinds of business decisions that must be supported by BI, which in turn is a function of the type of business process within which the BI will be deployed. Business decisions are generally classified as strategic, tactical, and operational, although in practice the distinctions can be blurred. These classifications tend to imply differences in

Table 7-2
Sample subjects and techniques for business analysis

Business decisions also take place in the context of business processes, such as management processes, revenue-generating processes, and operating processes. Depending on the process and whether the decisions to be made are strategic, tactical, or operational, different combinations of business information and business analyses will be appropriate. The general relationship between business decisions and business processes is shown in Figure 7-2.

At the core of Figure 7-2 is a highly simplified representation of value chain business processes, that is, supply chain processes, operating processes, customer relationship processes, management processes, intra-company processes, and support processes. In the context of these value chain processes, companies make a variety of decisions, shown in the arrows surrounding the core business processes. To illustrate, manufacturing companies make long-term strategic decisions about supply chain and operational processes, such as the number and location of plants and distribution centers and investments in information technology (IT) to enable planning and operational collaboration with suppliers. For a service company such as a bank, the long-term strategic decisions about supply chain and operational processes might address the question of whether to outsource check-clearing processes. In general, value chains differ by industry and product, and companies occupy different positions in those value chains. That said, the value chain construct is very useful for stimulating thinking about BI opportunities.

Faced with the many combinations and permutations of business information, business analyses, and fact-based decisions that BI can deliver or support, we need a general framework that delineates the major types of BI opportunities that can be considered for inclusion in a given company’s BI portfolio. With such a framework, we can further stimulate our thinking about potential BI opportunities as we conduct the BI opportunity analysis using the techniques described in Chapter 2. Table 7-3 presents some of the major types of BI opportunities on which companies in a variety of industries are capitalizing.

FIGURE 7-2 The relation between business decisions and business processes.

What we see in Table 7-3 is that BI opportunities can be segmented into three major business process categories: management processes, revenue generation processes, and operating processes. Within these major business process categories, proven BI opportunities are aligned with key business subprocesses, such as forecasting, customer segmentation, and order processing. Although the exact way that a given company competing with a certain business design in a specific industry might design its BI applications will vary, the typical BI opportunities shown in Table 7-3 cut across industries. Furthermore, although the business vocabulary of industries might vary, the need for management processes, revenue-generating processes, and operating processes is a constant. Accordingly, we believe that the typical BI opportunities shown in Table 7-3 are a useful place to start when determining the specific BI opportunities your company should pursue.

7.3 Business Intelligence for Management Processes

Building on the information presented in Table 7-3, we see that BI is commonly used to improve basic management processes—be they strategic, tactical, or operational in orientation. The business information and business analyses used for fact-based decision making range from information about the current status of the enterprise or its subunits, to historical information about past business performance that can be used to predict future demand and guide process and quality improvement efforts. Overall, the goal of the entire set of management processes is to ensure the long-term success of the company, and BI has a proven track record of helping companies make better decisions toward that end. To illustrate this point, we will provide highlevel descriptions and/or brief examples of some common ways that BI is used to improve management processes. We will describe two such uses in depth: balanced scorecards and management accounting.

Table 7-3
Major types of business intelligence (BI) opportunities and their relation to categories of business processes

7.3.3 Business Intelligence and Management Accounting: Improving Operational and Financial Performance

As Johnson and Kaplan (1987) elegantly describe in Relevance Lost, managerial accounting evolved in the early days of the industrial revolution as a tool for measuring and managing resource consumption, outputs, and productivity at the operational level of the firm. Early management accounting systems were devised by industrial engineers, and the business information these systems provided was essential for managing operational and financial performance.

The principal type of information provided by early management accounting systems was standard cost information, which was used for planning and controlling the productivity and efficiency of internal processes. Other key attributes of early managerial accounting systems were that (1) they were devised to provide the kind of information required to manage operations in the specific industry in which the firm operated; and (2) they combined financial and nonfinancial information.

An example used by Johnson and Kaplan to illustrate this was drawn from the railroad freight shipping industry, where a basic productivity measure was cost per ton-mile. Cost per ton-mile could be compared with revenue per ton-mile to calculate contribution margin, from which period costs and other indirect costs could be subtracted to determine how the business was doing overall. Furthermore, this information could be used by operations managers to plan, control, and improve efficiency. The combination of financial information and nonfinancial information that was specific to the industry, and type of operation was directly relevant to the operations management task of optimizing cost, time, asset utilization, and service.

Today, most observers agree that management accounting information derived largely from modern financial accounting systems is obsolete. A review of the management accounting literature finds this theme time and again: see, for example, Cokins (2001), Kaplan (1988), and Cooper and Kaplan (1991). Our discussions with operating managers in a variety of industries reinforce this finding. For example, during the course of our work with a $7 billion service organization to modernize and increase the relevance of its management accounting system, we conducted dozens of interviews at all levels of the organization, and we identified the need for the following:

• Information about the direct costs of delivering each of the different types of services offered to customers

• Cost information that is specific to each of the 1,500 field offices, as opposed to average cost information derived from aggregated financial accounting data

• Cost-to-serve information that allowed for cost comparisons by service delivery channel, that is, teleservice center channel, Internet channel, and field office channel

• Activity-based costs for assigning the costs of indirect activities based on the true resource consumption characteristics of the service lines that benefit from the indirect activities

• Cost-to-serve information that allowed for comparison by type of customer

• Full cost information for budgeting and pricing purposes

• Relevant cost information for process improvement and capital budgeting uses

As with many organizations, the current system of allocating expenses found in the general ledger (G/L) system down to service lines, channels, customers, and so forth generates cost accounting information in which the organization’s managers have no confidence. And yet the managers use that information because they have nothing better.

In addition to lacking cost information, the organization also lacks the information to systematically, consistently, and routinely perform fundamental operations management tradeoffs between asset levels, costs, process times, quality, service, outputs, and backlogs. In other words, the organization, similar to many organizations, does not have an integrated set of facts about organizational performance from which it can model, project, and analyze resource consumption under various customer-driven operating scenarios.

The commonly found deficiencies in cost information and the lack of integrated information about the other key operating variables of interest point to the need for expanded functionality for managerial accounting systems and for an expanded role for BI. To illustrate the differences between where we are and where BI can take us, we first examine traditional managerial accounting systems and the management tasks they support, as shown in Figure 7-3.

Figure 7-3 shows a representative managerial accounting framework, which we would note is concerned only with costs and not with the broader set of financial and nonfinancial information found in early managerial accounting systems and much needed today. The graphic shows the commonly used distinction between cost measurement and cost information uses. For cost measurement, the typical approach is to draw information from the G/L system and allocate it into a cost accounting model, which might be a standard cost model, a project cost model, or a process cost model, depending on the nature of the business. The costs thus allocated are then reported out on a period-by-period basis during the accounting year, with the reported information being aggregated or disaggregated according to a selected taxonomy of cost objects, such as organizational units, product lines, customers, geographic regions, sales territories, or other objects of management responsibility and attention. The reported costs can then serve as raw materials for various cost information uses, such as planning, budgeting, and performance measurement, as shown in the right-hand box near the top of the graphic. Figure 7-3 also shows some of the commonly encountered issues associated with traditional cost accounting systems, as reported in the literature and as we have encountered in practice. To move beyond these limitations, we need to provide a modern conceptual framework for managerial accounting information, such as the one shown in Figure 7-4.

FIGURE 7-3 A representative managerial accounting framework.

The modern managerial accounting information framework (MAIF) encompasses a broader set of information than a traditional cost accounting framework, including information about costs but also information about assets, quality/service, time (e.g., cycle time or process time), and outputs. Further, the MAIF is built on a single integrated source of managerial accounting information, which for sake of discussion we have simplified and shown functionally as a BI environment. Properly designed and built from proven BI products, the MAIF can serve as a relatively inexpensive but robust source of managerial accounting information that can be used to

FIGURE 7-4 The modern managerial accounting information framework (MAIF).

1. Improve management processes, such as planning, budgeting, control, and performance improvement

2. Improve specific key business processes that impact revenues and/or costs, such as the “customer order to cash” processing cycle

The BI products from which the MAIF can be constructed are proven and widely adopted. So are the management and technical methods for doing so.

With a customized MAIF in place, an organization is then positioned to truly leverage managerial accounting information to improve revenues, reduce costs, or both. This can be accomplished by deploying BI applications and incorporating their use into the key management and business processes that drive revenues and/or costs. An example of the range of such applications is shown in Figure 7-5.

Here are some examples of how a MAIF based on BI technologies and methods can be used:

• Western Digital is a global manufacturer of computer hard disk drives with annual sales of more than $3 billion. Western Digital uses BI technologies to better manage its inventory, supply chains, product life cycles, and customer relationships. BI enabled Western Digital to reduce its operating costs by 50% from their previous levels, thereby enabling increased profit.

• Capital One, a global financial services firm with more than 50 million customer accounts, uses BI to analyze and improve the profitability of its product lines, as well as the effectiveness of its business processes and marketing programs.

• BankCo is one of the largest banks in the world. It has more than $500 billion in assets, more than 10 million individual customers, and more than 1 million business customers. BankCo uses BI technology to better understand the major drivers of the profitability of its loan business, for example, default risk, and to better segment its customers on the basis of lifetime value so that it can offer targeted, differentiated service levels and pricing.

• TransCo is a leading manufacturer. Its parts supply operation provides more than 4,000 authorized dealers with service parts. It has more than 2,000 suppliers, hundreds of thousands of parts, and more than 1 million stock keeping units (SKUs) in its North American distribution facilities. It uses BI technology to improve order fill rates (improve customer service) while reducing safety stock levels (improve asset utilization).

FIGURE 7-5 How the modern managerial accounting information framework (MAIF) can support resource management applications.

Tip

Modern managerial cost accounting is another area where BI can make a surprising contribution to an organization’s effectiveness. One way to gain top management support is to show how BI can provide a far higher quantity and quality of managerial accounting information than can be obtained from current methods.

There are many other examples of how leading companies in a variety of industries are using enhanced managerial accounting information—delivered by BI tools—to drive improved profitability.

7.4 Business Intelligence for Revenue-Generating Processes

Some of the earliest uses of BI were driven by the desire to improve revenue-generating processes. Simply put, companies wanted to understand their customers better, retain their best customers, and sell them more products or services. Achieving those goals can be relatively straightforward in some business-to-business contexts, for example, if a company has a couple dozen key customers that drive a large percentage of revenue. However, companies that have millions of customers and that annually process tens or hundreds of millions of business transactions with these customers often can’t see the forest for the trees. Many such companies can’t identify their most profitable customers, they don’t know when profitable customers are about to defect, and they can only guess at what direct marketing offers will be most attractive to which customers.

To address these gaps in understanding customers, companies in such retail-oriented industries as consumer product retailing, telecommunications, and financial services were early adapters in using BI to sift through the data about the millions of customer transactions to better understand what drives revenue. By building on the information presented in Figure 7-3, we see that BI is commonly used to improve revenue-generating processes such as market analysis, customer segmentation, campaign management, advertising, channel management, customer relationship management (CRM), sales force management, and pipeline management. To illustrate this point, we will provide high-level descriptions and/or brief examples of some common ways BI is used to improve revenue-generating processes. Some of those ways are

For companies with millions of customers, perhaps thousands of points of sale, and perhaps multiple channels, such analyses clearly depend on access to relevant business information and appropriate analytical tools, that is, on access to BI. With a well-architected BI environment, marketing analysis can be done in near real-time to see current and long-term revenue trends and to understand the underlying drivers of revenue growth. With appropriate tools, data about hundreds of millions of individual transactions can be mined to answer the fundamental marketing questions posed above. Armed with better information, companies can be more effective in attracting new customers, retaining profitable customers, and achieving a sustainable revenue portfolio. Furthermore, they can understand the relationship between channels and profitability and introduce incentives for customers to use the more profitable channels. In general, experience in a range of industries has shown that one of the most profitable uses of BI is in better understanding the relationship among customers, products or services, and revenue generation.

Beginning in the late 1970s, psychographic segmentation sought to group customers by such potentially common characteristics as personality, leisure activities, and values. It became common to associate zip codes with psychographic profiles, with groups often labeled with catchy phrases as “pools and patios” or “shotguns and pickups.” With all of these a priori bases of segmentation, the connection between association with the group and actual purchasing behavior was not clear. To overcome this gap, marketers began to investigate differences in customer behavior as the basis for segmentation, initially with focus groups and market pilots and, in the past decade, by using BI tools and techniques. Specifically, by mining data about millions of individual customer transactions and marrying such information with traditional demographic, geographic, and psychographic information, companies have been able to group customers by purchasing behavior and to understand the relationships, if any, between purchasing behavior and, for example, demographic variables and product characteristics. These BI tools do not replace traditional segmentation and market research tools: they simply provide powerful new tools that work well with existing tools to help companies define narrower customer segments, understand the needs and values of those segments, create products and services that better respond to those needs and values, and develop more selective and effective ways to reach and acquire new customers and/or expand business with current profitable customers.

Except for enterprises that are completely vertically integrated, a rarity in most industries these days, companies face strategic decisions about what channels to use and which partners to use within a given channel. They also face the task of evaluating channel and/or channel partner effectiveness over time. All of these channel management tasks can be made more effective through adoption of BI. For example, we talked earlier about using BI for marketing analysis, from which we gain insight into how much revenue comes from which channels; through which channels volume is increasing, decreasing, or holding steady; and how different products fare in different channels. If we can integrate that information with appropriate channel cost information, we can determine which channels are most cost-effective for us as a means to margin optimization. By use of BI, we can also assess channel partner performance. For example, suppose a company uses different distributors in different parts of the country. With appropriate BI tools and techniques, we can assess their relative performance in terms of revenue growth and customer service. All of these BI opportunities, and others not detailed here, can enable more effective channel selection and management, which supports more effective revenue generation and growth.

7.5 Business Intelligence for Operating Processes

If we look at today’s operations management challenges, we see that the fundamental economic relationships among costs, time, service, quality, asset levels, outputs, and backlogs are still at the heart of what operations and supply chain managers must be concerned with if they are to optimize profits. The need for relevant business information about operating performance is even greater when one considers the trend toward extended enterprises and competition based on the performance of constellations of organizations up and down the entire value chain. We observed this need when we were engaged by a $2 billion food ingredients manufacturer for a supply chain collaboration and optimization project with an international fast food chain. In that engagement, the goal was to quantify and model the relationships among demand, inventory levels, service levels, costs, cycle time, warehouse locations, and operating policies so that the supply chain links from the manufacturer to the retail level could be optimized and periodically rebalanced as conditions changed. We found that we could get some of the information needed through ad hoc studies. In other cases, however, the information simply wasn’t available, either from the manufacturer’s systems or from the third-party systems distributors who were the link to the retail level of the supply chain. Yet this is exactly the type of information today’s managers need to optimize operations and operating profits among the companies in the extended enterprise.

7.5.1 Common Ways Business Intelligence Is Used to Improve Operating Processes

In our earlier discussion in Section 7.3.3, we illustrated how BI tools and techniques can deliver business information and analytical capabilities to help companies optimize operating performance in support of their business designs. By bringing together multidimensional information about all aspects of operations, BI provides the tools needed to improve asset utilization, reduce cycle times, improve quality, improve service, and reduce costs, all of which contribute to improved profits. To illustrate this point, we will provide high-level descriptions and/or brief examples of some common ways BI is used to improve operating processes.

To frame the discussion, we are concerned here with the highlighted operating processes (darker boxes) shown in the simple value chain representation shown as Figure 7-6. The fundamental operations management challenge is to optimize the highlighted processes to support the business design and thereby optimize operating margins, a key determinant of profit. Here are some common ways you can use BI to improve operating processes:

FIGURE 7-6 Operating processes in the value chain.

• Cycle time reduction. From a customer perspective, the time it takes to fulfill an order is often a key determinant of customer satisfaction, whether in a retail context or a business-to-business context. From a cost perspective, cycle time impacts capacity and asset utilization and can drive fixed costs up or down. With more and more companies today operating in lean environments and/or competing on time, cycle time reduction is increasingly important. Further, cycle time affects cash flow and working capital requirements, and thus it impacts profits. Accordingly, using BI to reduce cycle time is often a key business opportunity. Cycle time reduction is effectively a dimension of business process improvement, and thus having accurate and specific information about the end-to-end process is a prerequisite for knowing where improvement opportunities may lie.

For example, many companies pay a lot of attention to the so-called order-to-cash cycle time: the total elapsed time between when a customer’s order was received and when payment was received for the order. That entire elapsed time is a function of the elapsed time for all the steps in the end-to-end process. By using the kind of detailed transaction data that is often available from ERP and other operational systems, we can determine where most of the time is taken. By using analytical techniques from the operations research field, we can then identify improvement strategies and make operating changes. Furthermore, we can use the BI environment to gauge the effectiveness of the changes by examining post-change transactional information to see if cycle time has been reduced.

• Risk reduction. In operations, risk can convert to costs, which affect profits. The risk may be in potentially obsolete inventory, in warranty repairs, in fraudulent transactions, in bad debt, in employee turnover, in supplier performance, or in any number of other places. Because of the cost impact, being able to identify and quantify the likely economic impact of risk is an essential precondition for risk mitigation. As with many other BI opportunities, the business information needed to support informed risk analysis techniques can often be found in detailed transactional information.

For example, a very popular BI application in the credit card business is fraud detection, whereby a customer’s previous card transactions are examined by using data mining techniques to detect his or her purchasing patterns. This information is used to create a profile so that future transactions that appear to be out of the norm for an individual can be flagged and disapproved, thus reducing fraudulent transactions under circumstances as credit card theft. As another example, a major automobile manufacturer tracks warranty repairs by make, model, part, mileage, and so forth as a means of identifying the costs of quality problems. Armed with such information, the manufacturer can initiate value engineering changes to parts that show a higher propensity to cause warranty repairs. In general, by sifting through detailed transactional history, companies in a variety of industries have identified and quantified risks and then systematically reduced them, thereby reducing costs and increasing profits.

• Quality improvement. As with cycle time, quality improvement is effectively a dimension of business process improvement, and thus having specific business information about process quality, product quality, and/or service quality is a precondition for being able to improve quality. If we refer to Figure 7-6 and the value chain processes it depicts, we can say that the quality of every shaded process area can be the subject of quality improvement efforts. For example, in the firm infrastructure box, we might look at the invoicing process and track the percentage of correct invoices as a measure of quality. In the inbound logistics box, we might track supplier performance on a measure such as percentage of perfect orders, which we might define as all orders shipped on time, complete (no backorders) and billed correctly. By storing and analyzing detailed transactional information, BI can deliver essential information for identifying, measuring, and quantifying the economic impact of quality issues. This is a key step to improving quality and thus avoiding the cost of sub-par quality, thus improving profits. Brother International, for example, uses a data warehouse and BI techniques to glean information about product quality problems from its technical support calls. It then uses this information to improve both its products and its technical support.

• Service level improvement. Service level is defined differently under various circumstances: it may be simply a function of cycle time, which we discussed above; it may reflect more than one dimension of operational performance, such cycle time and quality; or it may be expressed as product availability to meet demand. In all these cases, we are concerned with being able to measure and manage service levels, and once again, we can use BI tools and techniques to analyze detailed transaction records to discern service level performance and identify improvement opportunities.

• Asset reduction. One of the principal ways that BI can be used is to reduce assets, particularly inventories of finished goods in product industries, inventories of operating supplies in various industries, and property, plant, and equipment in various industries. All of these opportunities are related to using BI to better understand demand and demand variability, or to improving process efficiency and thereby reducing capacity requirements in relation to level demand or enabling capacity to stay level in relation to increasing demand.

• Purchasing. In many industries, the percentage of the final product cost contributed by purchased raw materials, subassemblies, or finished goods is increasing for a variety of economic reasons. With purchased goods being a larger determinant of costs, many large enterprises need to maximize their purchasing power on a global basis. That said, many such companies have lacked a central repository of purchasing history, and thus they often do not know how much of a given product was purchased by their various operating units. Accordingly, strategic purchasing programs, which are generally organized by commodity, have lacked essential information to optimize their purchasing power. By using BI, this barrier to reducing costs can be overcome, as has recently been accomplished by a global automobile manufacturing company.

• Order processing. One view of order processing is from the cycle time perspective. In that vein, we can use BI to analyze and improve the process and reduce the cycle time, as we discussed above under the topic of cycle time reduction. Another view of order processing is from a customer service perspective, wherein we want to order better service to our most important and profitable customers. From that perspective, BI about customer profitability and/or customer lifetime value can be incorporated into business rules that prioritize order fulfillment to favor our best customers.

For example, in one business-to-business context in which we have worked, manufacturing scheduling was done on a first-in, first-out basis until it was determined that numerous small orders from infrequent customers would sometimes result in large orders for key customers being delayed. By using BI to identify its most profitable customers, this company was able to define business rules that could automatically be implemented by its manufacturing scheduling system, thus avoiding the costs of expediting large orders.

• Benchmarking and process improvement. Effectively, BI is a generalized process improvement tool, as the discussions above have indicated. An added wrinkle in the operations realm is benchmarking, whereby a company uses BI to analyze its processes along such key dimensions as time, cost, and quality and then compares its results to benchmark data. Benchmarking is a popular approach in operations, and some leading BI tools include reference databases of benchmark data.

7.6 Summary: Important Ways Business Intelligence Can Drive Profit Improvement

We have seen in this chapter that the way specific companies compete in their industries drives how BI can be used to improve their profits. We have also seen that there are many different possibilities for using BI, which creates some opportunity for innovation and competitive advantage. That said, we can use BI opportunity analysis to identify the best BI opportunities for a given company.

To stimulate your thinking about opportunities that may be relevant to your company, we have provided a wide range of examples of how BI is being used today in companies around the world. Whether your company wishes to innovate with its use of BI or simply keep pace, the BI opportunities described in this chapter are a good place to start. The bar has been set high in many cases, but not so high that BI cannot deliver competitive advantage and/or increased profits to companies that leverage what has been learned to date by successful BI innovators and apply appropriate methods for BI design and development.

7.7 Key Points to Remember

7.8 Think Tank