Kimball University: Building a Foundation for Smart Applications
Off-the-shelf apps may offer built-in analytics, but the best approach to supporting operational decisions is to rely on a solid data warehouse that cleans, integrates.
Enterprise Applications such as ERP and CRM systems are increasingly available with integrated analytic capabilities aimed at automating and improving operational decision-making. That doesn't mean you can get away without the solid foundation of an enterprise data warehouse and business intelligence (DW/BI) system.
No matter how customizable these applications may be and how extensive the analytic capabilities, your implementation team will still face the problem of integrating and cleaning data from multiple sources. Unless most or all of your transaction systems are part of the same integrated suite, tightly coupled with your smart apps, you'll eventually do much of the work that would otherwise be required to build DW/BI system.
The best way to move beyond information silos and support multiple smart apps is to build an enterprise DW/BI system. You'll gain integrated and conformed information from across the enterprise, granular detail with attributes and hierarchies that can feed any smart application, and rich history and change-tracking capabilities that can help you better predict customer behavior in any business process.
Organizations have two overarching drivers for their data warehouse/business intelligence (DW/BI) system: inform strategic and tactical decision-making, and automate operational decisions. The first goal is classic decision support: executives and analysts use information to run the business. With an excellent data warehouse, executives know what's going on in their business and markets, and can steer a course for success. The second goal can be realized by smart enterprises with enhanced operational systems that make use of data to run more effectively based on informed suggestions or predictions.
The smart enterprise weaves the information platform into the operational application layer. The DW/BI system can add value wherever operational decisions could benefit from a historical context. Examples range from setting order-fulfillment priorities based on customer value and personalizing product suggestions, to improving fraud-detection applications and manufacturing quality monitoring systems.
Decision support is great, but a smart enterprise is even better. The smart enterprise is not new: We've talked about and even implemented such systems for decades. Even the oldest data-warehousing architectural diagrams include a feedback- or closed-loop arrow. What's new is that the smart enterprise is moving into the mainstream, through both packaged analytics and custom applications.
The best way to get to the smart enterprise is to build an enterprise DW/BI system: an information platform that cleans, integrates and conforms data from across your organization. With this solid foundation, the goal of a smart enterprise is much more attainable. We will help you reach the desired end goal by describing the architectural foundation of the smart enterprise.
We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.