JUL 10, 2009 3:51am ET

Related Links

Obama: Better Federal Data Quality, Availability within Year
May 23, 2012
Bloomberg Launches Data Management Service with PolarLake Buy
May 23, 2012
Dispatches from MIT CIO Symposium
May 22, 2012

Web Seminars

The Big Deal About Big Data Governance
Available On Demand
Getting Started with Big Data
Available On Demand
Transactions & Interaction: The Correlation of Structured and Unstructured Data
Available On Demand

Overcoming Data Silos

Print
Reprints
Email
July 10, 2009 -- Every enterprise has business information that spans multiple data silos, but inconsistent, duplicate data scattered across various silos hampers effective decision-making. According to the Burton Group, a step-by-step program using MODS (Methodology for Overcoming Data Silos) projects can catalog and clean up data that resides deep within existing silos so businesspeople can readily perform cross-silo joins of important enterprise information.

Solutions built with MODS always reuse existing software and silos. MODS developers concentrate on data management more than on software development, and when they do software development work, they use nontraditional and relatively simple development tools 

Each MODS project results in the definition of a few relevant and important enterprise data types, as well as a registry that tracks and reconciles the instances of those data types in the disparate silos throughout the enterprise. 

A benefit of MODS is that the enterprises will learn how to bridge silos in an inexpensive and expeditious way. But because MODS is new and somewhat experimental, risks exist in terms of learning curve, the introduction of new technologies, and potential disruption in the IT staffing mix due to the types of skills required. 

MODS projects are centered first and foremost on data management, thus staffing may need to be tilted more toward data management professionals than developers.  In the early stages of a MODS implementation, the process of retrieving relevant information from disparate silos is a somewhat manual process. In the latter stages of MODS projects, businesspeople are able to meaningfully combine structured and unstructured data on any relevant business topic. Businesspeople can perform searches on enterprise data using a browser.

MODS is a stepwise approach that works best with a closely related set of data types. The best MODS project is narrow in scope and specific to a particular business topic. IT groups will need to adopt MODS gradually and follow five main steps: 
  • Step 1: Model/Understand the Relevant Data
  • Step 2: Identify, Reconcile, and Clean instances between silos
  • Step 3: Classify
  • Step 4: Convey (Carry, Deliver, Transmit, Express) the Information
  • Step 5: Control

"MODS projects enable enterprises to get more out of their existing systems by improving the information within those systems and making it more available to businesspeople," Lyn Robinson, the author of the Burton Group report, told Information Management. "MODS projects are low-risk, low-cost and high-return projects, and in today's economy, enterprise IT departments can hardly afford not to do them.

Burton Group proposes that  MODS offers a cohesive, step-by-step strategy for bridging data silos and delivering integrated information from within data silos to the businesspeople who need it.


For more information, visit burtongroup.com.

Julie Langenkamp is editor-in-chief of Information Management. She can be reached at julie.langenkamp@sourcemedia.com. Follow her on Twitter at @JulieLangenkamp.

Filed under:

Advertisement

Comments (0)

Be the first to comment on this post using the section below.

Add Your Comments:
You must be registered to post a comment.
Not Registered?
You must be registered to post a comment. Click here to register.
Already registered? Log in here
Please note you must now log in with your email address and password.
Twitter
Facebook
LinkedIn
Login  |  My Account  |  White Papers  |  Web Seminars  |  Events |  Newsletters |  eBooks
FOLLOW US
Please note you must now log in with your email address and password.