Informatica Enterprise Data Management

I teach data warehousing and business intelligence courses at Northeastern University's Graduate School of Engineering. It provides me a fresh perspective on what we are doing in the industry. This is because unlike when I am talking with customers, analysts and vendors, my graduate students are always asking "WHY."

They ask because they do not have our industry experience or background. Yes, I do need to take the time to explain a lot of things that I otherwise would not have to, but that is actually good news. The graduate students also have not been locked into a particular way of doing things because they see everything as a new experience. They have open minds.

It would be a good idea for anyone planning projects for next year to have an open mind, too.
One of the significant topics we discuss in class is data integration. Who needs it, what is it, how do you do it and when does it happen. We discuss breaking down the processes and data architecture that are used to integrate data from disparate and sometimes conflicting sources. We start by examining process and data before we talk products (even though the students are engineers and would love to talk products and technology).

My goal is to teach them that data integration is much greater than ETL (extract, transform and load) tools or a collection of SQL code that serves the same function. And once I teach them about how complex data integration is, I get them excited that technology and products have finally advanced to the point where we can use them to implement even the toughest and most challenging requirements.

Another take-away lesson is that the goal of data integration ultimately is providing the business with quality, consistent and timely information to enable them to do their jobs.

Why does this matter to corporate types like you?

I've consulted to many businesses that have been developing and deploying data integration solutions but are only barely tapping what a data integration suite can provide. When planning for next year or simply for your next BI/ DW project, it’s time to critically examine if you have gone beyond the ETL aspects of data integration.

Sure, many companies have been very successful using just ETL processes for very complex sourcing from disparate sources with significant data volumes. They have been able to create enterprise data warehouses to feed their reporting, BI and CPM applications. But in the majority of cases companies need more than just ETL.

Now that you're thinking about 2008, it's a great time to take an inventory of what data integration processes you have not yet adequately addressed. Often enterprises still need to expand their data quality and cleansing processes; leverage data profiling and modeling capabilities; and, implement metadata management to provide true transparency (auditability and traceability) from data source to information consumption.

Now is as good a time as any to build on your ETL success and expand into more complete data integration for greater return on investment of your company’s data assets. As an example, while your enterprise is planning to implement SOA to integrate its applications, why not step up and get data integration services on the drawing board? You should leverage your existing ETL processes into data integration services. After all, the most significant business ROI comes when you implement data integration services to enable business people to get the information they need anywhere and anytime in the enterprise.

(While you're planning for next year, be sure to take Informatica's free online Enterprise Data Warehousing Assessment test.)