The BI Justification Forest

Build a justification tree to help you determine the real reasons for that next business intelligence project.

• By Jeff Gentry
• 12/01/2001

Sometimes you can't see the forest for the trees—a common bit of uncommon wisdom.

Justifications for business intelligence projects are as varied as business strategies and vendor product claims. Many are good, others not. I'd say that the most common and absolutely worst justification ever concocted for data warehouses and data marts has been this one: "To provide better information to the business."

Shocked? Hoping a certain project sponsor isn't reading this?

So what's wrong with providing better information for the business? Well, nothing. Unfortunately, there may be nothing right about it either. If providing better information to the business doesn't materially help the business, the information has no value and the project that creates the information is unjustified.

With many businesses focused on cost-cutting measures, projects are justified by headcount reductions. If 20 people collect competitive information and sales figures in spreadsheets every week, and a new business intelligence capability can automate the entire process, costs are reduced only if the 20 people are removed from the payroll. Sometimes this happens, but more often those staff members are reassigned to other tasks. Is the project still justified? It depends on whether or not the value generated by the 20 people is materially helping the business. If not, there's no cost reduction and the project that eliminates 20 person-weeks of labor is unjustified.

Other justifications include arming salespeople with more accurate customer information, supporting product development efforts, streamlining supply chains, enabling more effective service delivery and providing better insight into financial status. Any of these goals may be reasonable justifications, but they're all justification trees. Unless considered within the perspective of the whole forest, the enterprise can't be certain that a supporting business intelligence project is justified.

Valid justification trees grow in four parts of the justification forest:

• Enhance revenue.
• Reduce operating costs.
• Gain competitive advantage.
• Increase the value of the enterprise.

Any justifications for business intelligence projects should be placed in perspective against the four broad reasons for business intelligence. Every extract, transform and load (ETL) process, every information access application and every element of every table should be justified against one or more of these giant groves of the justification forest.

Enhance Revenue
Simply put, enhancing the revenue of the enterprise brings more money in the door. Increasing revenues is often a business goal even if it doesn't increase profitability. In fact, many businesses will sacrifice profitability in order to enhance revenue during periods of rapid growth. The volume of revenue is proportional to the opportunity for future profitability when combined with reduced costs. Consider the impact of reducing the cost of generating $1 million in revenue by one percent by improving supply chain efficiency. The impact is $10,000. However, if revenue is enhanced to $1 billion, the same process change reaps savings of $10 million.

Even if enhanced revenues aren't coupled with cost reduction, they have the potential for greater impact when costs are reduced in the future. Sometimes these cost reductions are only possible due to the increased scales of the business resulting from greater revenues. Revenue enhancement, even if independent of cost reductions, is a valid justification for a business intelligence project.

Enterprises that are focused on enhancing revenues pursue business strategies that support this goal. Acquisition of new customers is usually at the top of the list. Within the existing customer base, sales and marketing can focus on cross-selling and upselling to bring more money into the coffers. Cross-selling is the practice of selling additional products and services to existing customers. All those inserts that come in the envelope with the credit-card bill are attempts to cross-sell. Upselling is the practice of convincing a customer or prospective customer to move from one product or service to a more profitable product or service. When the credit-card company calls to convince a customer to upgrade from a gold card to a platinum card at a higher annual fee, the company is upselling. A company may even increase revenues by raising prices even though it may result in fewer sales.

Business intelligence projects that help the enterprise implement these strategies are usually easy to justify.

Reduce Operating Costs
At times of economic downturn, or anytime that costs are growing faster than revenues, an enterprise is likely to focus on reducing the cost of doing business. To put it concisely, the business finds ways to use money more efficiently. Combined with enhanced revenue, the enterprise achieves higher profitability through cost reduction.

Enterprises focused on cost reduction pursue many strategies. Opportunities to reduce fixed overhead are sought out. Processes are streamlined. Greater priority is placed on negotiating better deals with suppliers. Employees are laid off or separated. The cost of interacting with customers can be reduced by encouraging them to interact through lower cost channels such as calling in orders instead of requesting a personal visit by a salesperson. Increasing the quality of products reduces warranty expenses. Improved cash flow and reduced inventory levels reduce the cost of capital.

The opportunities for cost reduction are varied, and business intelligence capabilities can often identify these opportunities. When the business takes advantage of these cost reduction opportunities, business intelligence adds value through cost reduction. If business just has "better information" without changing the way business decisions are made to reduce costs, there is no justification. A business intelligence investment must result in real, measurable cost reductions before the investment is justifiable.

Gain Competitive Advantage
Beating the competitors is sometimes harder to measure than cost reduction and enhanced revenues because external data is required to measure success. However, the value may be greater. The traditional measure of competitive advantage is marketshare; but in recent years, many enterprises are placing more focus on share of customer. Share of customer is the percentage of purchases of an individual customer gained by the business as opposed to a collection of all customers in a market.

Since it's often dramatically less expensive to sell more goods and services to an existing customer than to a new customer, this perspective of competitive advantage is important. Think of it as beating the competition one customer at a time.

Enterprises focused on competitive advantage may work to be more cost competitive. Alternatively, an enterprise can work to differentiate their products and services in such a way that they're more attractive to customers than competitive products. By combining a collection of business strategies to build better, mutually beneficial customer relationships, an enterprise can gain significant competitive advantages independent of all other factors.

Business intelligence capabilities are essential for the support of many business strategies to gain competitive advantage, especially in the realm of customer relationship management (CRM), a business philosophy that aligns business strategy, business culture, customer information and technology to build mutually beneficial relationships between an enterprise and its customers.

Increase the Value of the Enterprise
If we consider enhanced revenue, cost reduction and competitive advantage the three groves of the justification forest, then increasing the value of the enterprise is based on the potential future health and growth of the entire forest. Now that the age of "irrational exuberance" is gone and the dotcom bubble has burst, the traditional methods of valuing a business have returned. The prospects for an enterprise to enhance revenue, reduce cost and gain competitive advantage into the future are the strongest indications of an enterprise's increased value.

Businesses that are focused on initial public offerings (IPOs) or on being acquired want to increase the value of the enterprise. There are numerous examples of business intelligence capabilities contributing to the value of the business. This is an excellent justification for business intelligence projects.

Keep an Eye on the Forest
Like any forest, the justification forest is the composite of many trees. Data marts and BI applications are usually built to support focused business needs represented by individual trees, such as, "Reduce the risks associated with outstanding accounts receivables." However, every time we justify a project based on one of these trees, we need to validate a few things by asking these questions:

• Will the project enhance revenue, reduce costs or gain competitive advantage?
• Can any piece of the project be eliminated—even a single data element—without reducing the ability of the project to enhance revenue, reduce cost and gain competitive advantage?
• Will the project deliverable support current business plans?
• Will the business change as a result of the new BI capability?

By putting a project into the framework of the big picture, it's much more likely that the resulting capabilities will result in real, measurable, long-term business value. After all, this is the only true measure of success.