Today's editorial was originally released on Nov 27, 2007. It is being re-run as Steve is at the PASS Summit.

My guess is that most of you out there wouldn't necessarily classify your managers as "savvy", especially as it relates to data. I know that when I hear managers talking about "data quality", "data integrity", "ETL", or any other acronyms, I tend to cringe, expecting more work, with poor requirements, and likely unreasonable estimates.

But not all managers are that bad and when I attended the Micosoft BI Conference earlier this year, I was surprised to see so many business people, especially managers, there and talking about how much value they got from BI because it gave then more insights into their data.

And they almost universally talked about how important a strong data warehouse with high data quality is to the success of a project.

Data quality is important, but it takes an effort to ensure that you can achieve a high level of quality in your data, meaning that the data is accurate and represents what you think it represents. I caught this interesting article about 10 data quality habits for successful managers. It probably should be for successful organizations and not just managers, but it's a good guide for managers to be aware of. Without reading the article, I'd bet you could guess at what some of the items should be. They're mainly common sense, but they bear repeating and it's good to see them listed in together in a short article.

Data quality takes effort and just like programming, the earlier you can introduce checks and catch errors, the less expensive it is to maintain. However that doesn't necessarily mean that you should go all the way to the source. Putting in a huge amount of checks and filters in the input client might not be in your best interest.

Consider a salesman, trying to make a sale, entering data and constantly getting pop-ups and errors that force data entry to be exact. Can you imagine how frustrating this would be? And possibly how this might impact data quality? Can you guess at what percentage of people might get names entered as initials instead of misspelled names? Does someone need to be slowed down because they typed "Bbo?"

Enforcing data quality at the source might be better served with suggestions or filters that try to fix common mistakes or even batch up confirmations of suspected errors for someone to examine later. There are any number of ways to make this an easier process and still ensure data quality.

My advice is that you should tackle data quality as an ongoing part of your job. Make constant, continuous, and small improvements, build in checks and balances, and be sure you work with other groups and users to ensure the load is shared, and more importantly, easily integrated into the way they already do business.

And maybe I'll see you featured at one of the next BI events as a SQL Server success!

Steve Jones

The Voice of the DBA Podcasts

The podcast feeds are now available at sqlservercentral.podshow.com to get better bandwidth and maybe a little more exposure :). Comments are definitely appreciated and wanted, and you can get feeds from there.

The RSS Feed: or now on iTunes!

• Quicktime Podcast - 24.0MB
• Windows Media Podcast - 11.1MB
• MP3 Podcast - 4.9MB

Today's podcast features music by Everyday Jones. No relation, but I stumbled on to them and really like the music. Support this great duo at www.everydayjones.com.

I really appreciate and value feedback on the podcasts. Let us know what you like, don't like, or even send in ideas for the show. If you'd like to comment, post something here. The boss will be sure to read it.