Three Tips to Maximize the Value of Big Data

When IT leaders delve into a “Big Data” challenge, they discover that it is more complex than they realized. Enterprise architects, information managers, and data management and integration leaders often find that massive volume represents only one aspect of the problem.

Looking beyond managing sheer volume, certain deficiencies within Big Data infrastructures emerge, whether Hadoop-based, ADBMS-based, or a combination of both. Organizations seeking to maximize the business value of Big Data must effectively address these issues.

Here are three suggestions to help your organization maximize the business value of Big Data:  

1. Close the gap that exists between the business user and Big Data.

Current implementations require many of the same components needed for traditional data warehousing/ business intelligence (DW/BI) systems, including developing the necessary data schemas and SQL queries. These are (typically) not directly accessible by the massively parallel processing (MPP)-based distributed file systems, such as Hadoop, which routes data in batch mode. Together with the normal latency associated with data warehouses, this can lead to blind spots impacting real-time decision making.

2. Manage multiple sources at a time.

Data systems typically manage one source at a time. As a result, complex relationships between data types are often missed. Information that is managed and analyzed in isolation as a (very large) silo can yield only a subset of its potential insights, and may produce more questions than answers.  A complete informational picture that would be realized by integrating other relevant enterprise records is possible with Big Data.  Organizations must rethink their siloed views of big data, reframe their enterprise approach for access and analysis, and resolve problems using analytical methods that are made possible through unified data.

3. Develop contextual business understanding.

Unstructured content such as documents, email, web content, free-flowing text, SharePoint, call logs, and surveys, are neither classified nor analyzed to provide contextual insights. Most Big Data systems will assert that they “handle” data that is not structured; but they don’t perform text analytics or sentiment analysis on truly unstructured content. Contextual business understanding, which can hold the key to uncovering new business insights, must be developed.

A review of these concerns with existing Big Data infrastructures confirms the challenge reaches far beyond volume to encompass the full enterprise challenge. Instead of merely managing Big Data as its own separate silo, readily and systematically attain new data insights using the tips outlined above.

For a more in depth look at this topic click here to download “Completing the Big Data Picture with Unified Information Access” white paper.

Sid Probstein is the Chief Technical Officer at Attivio, an enterprise solution company that helps organizations identify trends, filter Big Data, implement analytics, and streamline costs and efficiencies. To learn more about its new Unified Information Access (UIA) platform, visit Attivio.com.