Data Integration Remains a Major IT Headache
Study shows that data integration is still costly and requires a lot of manual coding.
Remember when "integration" meant sneakerware
? You put in a floppy (USB today), copy something, walk it to another computer, insert disk and copy.
"They are circumventing the lack of integration by using humans to move information from system to system," Linthicum states. "As a result, information is often entered incorrectly, and data quality suffers greatly, including the potential of making critical errors."
What's surprising is that these organizations often already have data integration technology, but they're just not using it smartly, he adds. Health care organizations are using the tools in a limited, tactical way to connect two data resources together, rather than using it more broadly to link critical systems together.
"I would argue that this technology is strategic in nature, and really needs both a strategy and deployment plan to get the right data at the right time flowing freely between the healthcare provider's critical systems," Linthicum writes. "You might be losing thousands of dollars a day in lost productivity due to the lack of a good integration strategy and a good data integration technology provider."
It's easy to look down your nose at health care's backward IT practices, but I suspect health care's not alone in underestimating the need for a unified approach to integration.
Identify the problem in your hand and future growth aspects. If you do not have some heterogeneous complexity, there are good tools for less TCO. Eg: Did anyone use all the features of Informatica (other ETL tools) in their business? No one want to use Sledgehammer to drive (a) simple nail.
True enough. Not everyone needs a sledgehammer. And, obviously, some may be overkill in your situation.
But most integration solutions - ETL or otherwise - now come with features that may seem extraneous, but are actually critical to supporting an integration strategy. Support for data quality, data governance, meta data, master data - smart organizations understand these features are fundamental, not "nice to haves."
You have to wonder how many IT organizations - much like health care - are using these tools tactically rather than looking broadly at integration. I've talked to many integration experts over the years and, at a minimum, developing an integration strategy requires:
Are you wasting thousands of desperately needed dollars or more in lost productivity because you don't have an integration strategy and because you haven't bothered to take a unified, organization-wide approach to integration? If so, you probably shouldn't snicker at health care's sneakerware until you move into this decade with your own integration practices.