Integrate the Data

 
 
By Jake Freivald  |  Posted 2010-03-12
 
 
 

How to Use Enterprise Information Management to Increase Data Utility


"Water, water, everywhere...Nor any drop to drink." So said Coleridge's ancient mariner about dying of thirst on a languishing ship, surrounded by the salt sea. I feel that way sometimes about data. In lean economic times, we want to capitalize on every resource we can find but the data that surrounds us seems unusable. It's been estimated that 80 percent of vital business information resides in unmanaged repositories, making effective utilization nearly impossible.

Before I get e-mail messages from grammar wonks-always a danger when I start with a poetry reference-I say "utilize" instead of "use" deliberately. Before the data can be used, we have to bring it from its unusable state to a usable state. Just as we must desalinate seawater in order to drink it, we must increase data's utility before we consume it. Let's look at three tasks you can do to improve the utility of your data, as well as the three types of related projects that can help you in lean economic times.

Task No. 1: Clean the data

As with water, cleanliness improves things everywhere you use data. If you want to increase revenue by maximizing campaign effectiveness, you may want to segment your data better-but target selection can be tough when you have titles such as "Chief info officer" or "Integratoin mgr" instead of correctly spelled and standardized titles.

If you want to cut costs, you can start by deduplicating the mailing addresses of your catalog recipients. If you want to avoid the "float" of your billing, clean your address data to ensure that clients receive their bills as quickly as possible.

Integrate the Data


Task No. 2: Integrate the data

To make data more valuable, make more data available. By that I don't mean you should seek a "single version of the truth" through long and costly data warehouse projects. After all, people only care about their version of the truth, not the varying "single" versions stored in three or four "enterprise" data warehouses.

Instead, look for business priorities that would benefit from small-scale, real-time integration, and use best practices in interface design to implement them.

Task No. 3: Cross-reference the data

As you get better at identifying projects that will have a significant impact, you'll see the need for the same kinds of information over and over again (for example, patient or medical equipment information for healthcare and part or logistics information for manufacturers). By identifying systems of record with accurate and up-to-date information in them (the so-called "master data") and consistently referring to them in integration projects, you can increase efficiency, decrease programming time and improve quality in nearly every process in your organization.

This process, called master data management (MDM), is relatively advanced and shows up under other names such as customer data integration (CDI) as well. These three topics can be combined, with others, under the umbrella term enterprise information management (EIM).

Using EIM to Realize Immediate Impact


Using EIM to Realize Immediate Impact

Over the long haul, EIM can support customer service and regulatory compliance, help companies identify new marketing opportunities, improve processes, and even detect fraud. But in the short term, I've outlined three key strategies in which companies can use EIM to realize immediate impact, while laying the foundation for a larger, more robust information management system in the future.

Strategy No. 1: Delay ERP projects

Companies rely on ERP systems to streamline core activities, consolidate data, and enhance visibility into key operations. But the average cost of an ERP expansion can run as high as $1.5 million.

Instead of upgrading your ERP, consider small-scope projects that can achieve similar benefits. For instance, if you wanted better consolidated data, ask yourself, "What are the five most important questions that this ERP upgrade would help me to answer?" Then, to answer those questions, invest some money-much less than your ERP upgrade-in software and processes for data quality, integration, and business intelligence (BI).

Strategy No. 2: Automate processes to increase workplace productivity

According to the International Society for Automation (ISA), automating a process alone increases efficiency 1 to 2 percent annually. But enriching the same automated process with real-time information yields an additional 3 to 5 percent improvement.

So, find one valuable transaction within your company: maybe your ERP system's "insert purchase order" transaction. Then, publish an event that contains that transaction plus cleansed, standardized business data enriched with the latest data about that purchaser. Finally, have one application (for example, your CRM) subscribe to this event stream for real-time updates. This will minimize manual data re-entry and make a noticeable difference in staff productivity.

The transaction doesn't have to be from the ERP system, either. Publishing an event from the corporate Website or a business-to-business transaction, and then tying that event into other applications will eliminate manual data entry work. It will also enhance information accuracy, yielding higher productivity for users.

Strategy No. 3: Avoid complex BPM deployments

The average cost of a business process management (BPM) project can be between $250,000 and $500,000 in licensing fees, application and database servers, internal staffing costs and training expenses. To be successful, BPM deployments require business-level services, as in service-oriented architecture (SOA)-not consulting. Their interfaces use only business-level data, without reference to proprietary applications.

Consider implementing a few key services (maybe "check claims status" if a lot of applications need claims information) as a temporary measure. Then you can call them from the company's current applications, portals, BI and reporting environments, and other technologies. You'll gain a lot of efficiencies in the short run and you'll accelerate your BPM implementation down the road.

Conclusion

These three projects are small by design. They have to be so that you can implement them without spending a lot of money. But they focus on improving and using the data you already collect, which will save time and money in the short term while getting you ready for longer-term benefits as well.

Jake Freivald is the Vice President of Data Strategies for Information Builders (and iWay Software, an Information Builders company). Jake joined Information Builders in 1999 as a product manager where he supported cross-platform product development and deployment. During his career, Jake has held several managerial positions with Andersen Consulting and Prudential Life Insurance Company of America.  He also served in the United States Marine Corps as a Signals Intelligence Officer at First Radio Battalion. He graduated from Cornell University with a Bachelor's of Science degree in Electrical Engineering. He can be reached at jake_freivald@informationbuilders.com.

Rocket Fuel