Three Things the Data Team Wishes You Knew About Data Modeling

Loraine Lawson
Slide Show

Data Integration Remains a Major IT Headache

Study shows that data integration is still costly and requires a lot of manual coding.

Data modeling may not be high on your priority list of topics to study. And who can blame you - it's a deeply technical and some might say tedious issue. But data modeling is attracting more attention these days as a valuable tool for taming the chaos of data.


There are a few things the data team wishes you - the IT manager, CIO and even business managers - knew about data modeling, however. To save you time, here's the short list:


1. Data modeling - and data in general - should be handled separately from application development projects. Data modelers are big-picture thinkers - they spend their time ensuring that the data captured and processed benefits current and future business needs.


Application development is about solving an immediate problem as quickly as possible, Karen Lpez, (@datachick) a data modeling expert and the senior project manager at the management consultant company InfoAdvisors, explained during a recent IT Business Edge interview.


"That mismatch of goals for the work leads to a lot of problems as once data is captured, it's very difficult and costly to try to build quality back into it," she said. "In fact, in most situations it's impossible."


2. Data models are worth the extra effort. The way application development has traditionally happened has lead to data silos, each with slightly different variations on how the data's stored. Data models can change this, making it easier for all systems and applications to share information.


"Data models (and all the other parts of a strong data architecture) allow us to document the data as it exists in these disparate systems, then to compare and contrast the differences," Lpez said. "The more the models are used, the easier and cheaper we can build systems that work with each other."


3. Data models don't have to be "finished" to be useful or pay off. " A data model is never complete, but that should not deter using it to support development projects," Jonathan G. Geiger, executive vice president at Intelligent Solutions, Inc., said back in 2007. Geiger actually advocated for developing the data model with an application process, which Lpez says can create problems. But the bigger point here is that your data model is a working, living document and not something to keep locked up in the basement with, say, model trains. So put your data model to work, advises Lpez.


"Love your data," she adds, a slogan she uses often. "Data lasts longer than code, technologies, buzzwords and hardware. It's what we make decisions on and build the business on."

Add Comment      Leave a comment on this blog post

Post a comment





(Maximum characters: 1200). You have 1200 characters left.



Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.