Top Ten Best Practices for Data Integration
Use these guidelines to help you achieve more modern, high-value and diverse uses of DI tools and techniques.
Would you like to cut costs on data integration? Then use XML and take steps to streamline the transformation of XML to common data formats, recommends the Aberdeen Group.
And it's not just XML. You can also reduce costs by using hierarchical data standards such as JSON, COBOL records, Google Protocol Buffers and Apache Avro, which also are linked with lower integration costs, according to Aberdeen's paper, "XML and Hierarchical Data Integration: The Challenge, the Cost, and the Opportunity." The report is now available for free download starting this week, for a limited time.
Not surprisingly, this approach also saves time: The average employee spends four and a half hours every week tracking down information they need to do their job. But at companies using XML - coupled, the report is quick to add, with better methods of transforming the data to other common formats - and that number drops by 10 percent over 12 months.
So just how does one go about "streamlining and optimizing the transformation of XML (or other hierarchical data) to other common data formats"?
Aberdeen recommends you pick one data integration tool for all of your integration and transformation projects. No more separate solutions for specific projects or individual divisions. Just one tool to integrate them all.
That, it turns out, is the key to the real savings, resulting in:
And, at the risk of sounding like an infomercial but wait, there's more:
"Aberdeen estimates that these organizations could gain further cost savings by increasing the productivity and efficiency of their data integration processes, so cutting the time required to create and modify data transformations," the report notes. "These benefits typically accrue from the improvement in skills and talent that focusing on a single tool can bring, as well as the reuse of existing transformations and integration component libraries."
Recently, I argued that data management should get some props for the ROI of analytics projects. Sure enough, Aberdeen states that data management typically accounts for 42 percent of the total resources used during an analytics project.
Although most organizations - 71 percent - are already integrating XML, the task takes plenty of time and money. All but a handful of the survey respondents use manual processes to integrate external sources.
To better deal with the new integration demands, 34 percent of respondents say they'll add a tool that will automate the capture and collection of XML data. Forty-six percent currently have that capability, so that's a huge amount of companies buying tools over the next 12-24 months.
The full seven-page report includes a lot more details and fun statistics on how companies handle hierarchical data, as well as a page of recommendations on how you can cut the time and money spent on data integration, depending on what you currently do. And, given that it's free, it's certainly a worthwhile weekend read.