Golden Databases – Caveat Emptor

I was very interested to read a great post by fellow Irish blogger Damien Mulley in which he wrote:

…Most companies have massive databases of customer details that are sitting there, gathering dust. Why not work on those databases and poll your customers …

The context of Damien’s comment was a larger piece about using key assets in your organisation to drive up business or drive down costs. Damien rightly points out that information (on customers) is a valuable asset that most companies simply don’t have working for them.

He references the Obama campaign, which many hail for using all the bells, whistles and tweets of Web2.0 but which was ultimately driven by good management and application of the campaign’s information assets.

He’s perfectly correct. I wrote about exactly that topic here (November 2008) and have touched on it in other posts and articles. Unfortunately, Damien’s “Golden Database” needs to come with a big Caveat Emptor.

Simply put… if your customer data has been “gathering dust”, you need to be prepared to find significant problems with the quality of the information in that database.

This is a topic that I’ve written about before for the Small Firms Association magazine “Running Your Business” back in 2007 (I’ll be posting the text of the article over here shortly).

Basically, information is just like any other asset (more or less. It does have some special and unique properties that would take to long for me to get into here). Just like any other asset, you need to Care for it and Feed it. You train staff, you service company cars and equipment, you keep your offices clean etc. (Tom Redman goes into this in length in an article on the IAIDQ website from 2007).

However, in most organisations, data is not properly cared for or fed. Often it is regarded as a by-product of sales or other activities and is not seen as a key input into other processes or a crucial treasure trove of knowledge to drive competitiveness or simply just save your business.

In my last post, I quoted research from Gartner Group which  found that

Three-quarters of the respondents consider data quality problems a constraint on, or a barrier to achieving, business success. Even so, only 41% of their organizations have a formal improvement program — the rest are doing nothing formally to improve matters.

That “constraint on business success” is the source of the caveat emptor I am advising. Common problems that may affect your “Golden Database” include

  • Duplicate or “fragmented” customers (customers created twice or multiple customer records for different departments or people in the same business). This could lead you to chase business from people who (for example) you are actually suing for non-payment of bills.
  • Incomplete or missing data (e.g. the “.com” missing from an email address, an area code missing from a phone number, a name missing from a contact record).
  • Inconsistent data (i.e. the facts don’t add up. A “Mr” with a gender of “Female”, a “Key Customer” flag but zero sales revenue etc).
  • Inaccurate data or Just plain wrong data (for example, your primary contact has left the firm, or has died. Or the business has ceased trading, or was never actually a business or a customer of yours. This is also a Data Protection requirement).
  • Unwanted “narrative” in fields (e.g. comments written into address fields etc. I have actually seen this happen and result in litigation and lost customers.)

As for the relationship between Information Quality and Data Protection, it is simple. By actually managing your information quality and caring and feeding for your information assets you end up adopting practices that ensure the accuracy and “up to dateness” of your data. If you want to know more about that, check out this webinar from the IAIDQ, or check out my presentation below (delivered at conferences in 2009).

The good news? If you heed  the caveat emptor warning and put in place processes to act on feedback you receive about the quality of your data you can polish your golden database to remove any tarnish. This, unfortunately, is a low added-value ‘scrap and rework’ task. (I wrote about why this was not a sustainable approach to ensuring high quality information here back in 2006).

Even better, you might invest in tools (some of them are free services) to profile the quality of your data to let you know just how much emptor you need to caveat. This would let you ‘ring fence’ particularly cruddy data so your customer doesn’t see how much you really cared during the boom times (when  you weren’t caring for or feeding your information asset). It would also save you from spending money on futile mailings or wasted phone calls (those things cost money don’t you know!)


All that glitters is not necessarily gold.

But if you accept that fact, and accept that any asset that has been left gathering dust will likely have issues in operation (like a clapped out banger company car) and that you will need to invest some time and effort in your “care and feeding” to improve the quality of your information asset, then you are well placed to reduce the “information quality friction” that drags on your business strategies and reap the competitive benefits of higher quality information in your business.

If you don’t…. caveat emptor.

My Presentation