What Are You Measuring?Before you can create a useful analytic application, you must first analyze the business effectively. A method of business assessment developed two decades ago can form the foundation for the most cutting edge of today's strategic IT applications
by Creighton Lang Continued from Page 1 For instance, most retailers spend significant amounts of money on marketing, making this activity both a high-cost function and a large revenue-generation opportunity for business analytics. The store dimension is necessary for a number of marketing analytics, as analysts slice-and-dice sales data and promotional effectiveness by store attributes such as square footage, geographic region, amount of time spent, and item selections. Attributes of the store dimension are also needed in other nonmarketing analytic opportunities such as profitability measurement, workforce analytics, or procurement efficiency improvement. Since the store dimension will be reused in future data marts, the return on the analytics to support the marketing process is increased. Because marketing needs a number of these shared dimensions, it's often a good starting point for retailers. As you deploy the architected data marts, you eventually create analytic capabilities in all your most influential business processes and will, in essence, incrementally deliver the enterprise warehouse capable of supporting true corporate performance management. I offer three examples that illustrate how some enterprises are using this framework to maintain competitive advantages in today's slower-growth economy. Data warehousing can be used:
Single-Activity OpportunitiesThe most mature analytics developed today focus on a single activity or business process. Included in firm infrastructure activities are business functions such as management, finance, and accounting. Consider the CFO at a manufacturing company. He wants to reduce days-sales-outstanding (DSO), but must manage in a fragmented culture in which different departments oversee their own open receivables collection. By creating an architected data mart focused on this subject area, the organization can create multidimensional analysis opportunities by including dimensions such as department, shipping location, client, employee, product, and, of course, date and time. Individual managers are now able to first understand the current receivables patterns and needs and then work with the CFO to define new limits and targets. After organizational change management was complete, the BI solution subsequently enabled both executives and staff to manage by the numbers, identify where the most organizational benefit existed, and monitor (in some cases compensate) based upon compliance with the predefined targets. Combined with increased management attention, the firm was able to use the business analytics solution to reduce DSO from the high 50s to the mid 40s in less than nine months. Opportunities at Activity IntersectionIn the second example, a financial services firm wants to understand how customer service across various channels affects current and future sales and customer profitability. In this case, the firm is implementing analytics that seek to deliver insight into two unique value chain activities: marketing and sales, and customer service and support. By following the architected data mart approach, the company developed marts for analyzing the various aspects of the sales and marketing organization. Separately, data marts were developed to track customer service and support across Web, interactive voice response, call center, and branch interaction channels. Because they were part of an architected approach, a new analytic application using both sets of data was relatively simple to deploy; data integration was already solved. The new analytics help the firm optimize the degree of personalization based on customer profitability and historical channel use. Sales analytics were implemented that have increased cross-sell success rates because they take into account not only formal selling promotions but also interactions with the various service organizations within the company. Finally, the financial services firm is able to optimize staffing in the costly call centers and branches because it has better intelligence on how wait times detract from future sales. The company has been able to reduce full-time staffing without sacrificing profitable service and has reduced overtime by accurately predicting which marketing promotions will generate overflow call center activity.
|
 |
Most Popular This Week
IE Weekly Newsletter
Subscribe to the newsletter
| |
|