CIO Baby Steps for Cleaning up Dirty Data

Loraine Lawson
Slide Show

How to Sell Senior Management on a Data Warehouse

The real benefits of data warehousing are indirect: the ability for your company to make better, faster decisions that will save money and increase revenue.

Dear CIO: I shouldn't have to convince you that dirty data is a huge issue for most companies, most likely including your own. No, the real question isn't whether you need data quality, but rather what you're going do to improve data quality.


But if you want a few figures to chew on, try these:


I could go on and on (which, now that I mention, I have done), but you get the idea: Dirty data, bad. Data quality, needed.


So how do you begin?


The Data Warehouse Institute recently published five steps you can take to improve data:

  • Dedicate resources (meaning, more than one person) to maintaining data integrity.
  • Embed your analytics into business operations, because then you can add business rules to enforce rules about your data and automate some of the data cleansing work.
  • Don't force an overarching schema, which can create data errors when business users find the data won't fit into your schema.
  • Allow visibility into the history of the data, including where it originated.
  • Get your data out of Excel. OK - TDWI was more diplomatic than that, advising you to, "Think beyond Excel." But seriously, isn't it time to outgrow Excel, which just creates data silos and doesn't support things like auditing or automation?


The full article goes into more detail, of course, but this is only a tips list and not meant to be comprehensive.


Certainly, there are other, more detailed paths to data quality, whether that's by establishing a competency center or starting a pilot project in the area of greatest need.


But what if you really don't have a lot of time, staff or money to devote to data quality? Is there one, small baby step you can take to improve your data today, without committing to a larger project?


Fix your data collection forms, advises Data Quality Pro founder and community leader Dylan Jones.


"If I had a limited amount of cash (and most fledgling data quality initiatives fall into this category let's face it), I would sink my pot of cash into improving the most frequently accessed forms that your customers or data entry workers use to enter data," Jones wrote recently. "Identify some simple, quick wins to enable mistake-proofing by making them clear, concise, clever and co-operative."


It's a baby step, to be sure, but even small steps will eventually get you there.

Add Comment      Leave a comment on this blog post
Jan 20, 2014 3:06 AM Mathew Rotlen Mathew Rotlen  says:
Thanks for useful informative. To manage the company data and keep it accurate, it must require to cleaning up of dirty and worth less data. Reply

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.