Five Tips for Easier Data Governance
Five steps you can take to ease the trauma of starting data governance.
You'll often see an admonishment to "focus on the pain points" with data governance and its close cousin, data quality. And that's good advice, but how do you identify the pain points and how do you know which pains are really worth pursuing?
This is apparently a real challenge for companies. One IBM-sponsored study, cited recently in Information Management
, found that two-thirds of all companies are implementing or planning to start a data governance project within the next year and a half. But most also said their inability to communicate the value of managing data was a major obstacle for data governance.
I feel you rolling your eyes out there, but what I really respect about Hayler's writings is he always takes a very practical, quick approach to building a business case. As Hayler describes it, your business case is the process where you hone in on the smartest starting points for data governance and decide - yay or nay - if that path will pay off in real business value.
In other words, it's the way you hone in on the pain points that really matter.
As it turns out, data governance doesn't have to be this all-encompassing, massive project. You can actually reap big benefits from focused efforts. In practice, only about 20-30 percent of a company's business data is actually critical and strategic, according to a recent Information Management article on data governance. Target that data, and you'll find success.
How do you do that? Here are some ideas to help you find the pain points and make the case for data governance:
Look at the business processes and - gasp - talk to the business process owner. Hayler says business process owners can tell you about their bottlenecks and problems. It's then your job to find data's role in those problems and associate the cost to fix it.
Identify the most business-central, strategic data
, advises consultant Maria C. Villar and Cisco Systems marketer Theresa C. Kushner in "10 Steps to Effective Data Governance for SMBs
." The most important data is tied to your business model, so, for example, if your business is service-oriented, look for data issues in customer-related data. Manufacturing businesses, on the other hand, will be more concerned with inventory and product data.
Look at how data management or governance can help with a major business driver or company goals, Villar and Kushner also suggest. Mergers and acquisitions, business intelligence projects, business process re-engineering or other processes may provide a sort of on-ramp for data governance.
If that doesn't help, here's a list of situations where data quality problems can lead to major losses, according to Villar and Kushner:
- Bad billing data, which can cost millions in lost revenue
- Errors in contact information, such as the wrong gender, which can alienate customers
- Financial reporting data problems, which can lead to financial penalties, restatements and even jail time
- Data security breaches caused by inadequate data governance, which can lead to fines, bad press and angry customers
- Inaccurate sales forecasts, which leads to wrong earnings expectations
Once you've identified a data problem that you think could benefit from data management or governance, figure out the costs of the problem. Sixty-eight percent of companies didn't measure the costs of data quality, Hayler says, citing a 2010 Information Difference survey of 257 companies.
By measuring the costs of your data problems, you can better determine whether the costs of fixing those problems are worth pursuing. Usually, it will be, he adds; data is often so mismanaged, you can come up with "some surprisingly high potential savings," he writes.
Once you've identified a data problem where governance could create savings, you've not just found your pain point - you've found your business case.