Very good point! Taxonomy is to data profiling as good foot skills are to excellent football (soccer). One begats the other – its the chicken and the egg of data quality. Which comes first – a decent taxonomy or the effort of data profiling. I suppose that if we start with the business folks in a weak data infrastructure environment, we would purly speak about taxonomy. Given a somewhat mature data infrastructure or starting from IT, we would most likely want to begin our discussions on the interim data profiling results.

Either way, nice catch Prashant!

In your post “Approach to Start the Data Profiling” – I agree that if you are very familiar with the data and are equally talented in sql, then a db driven approach may be appealing. On the other hand, purchasing tools gives broader coverage of profiling ability. Like you are implying, you have to weigh the benefits of each to make the decision.

Take care,

~Scott Felten

I guess the missing part in the above list is Data profiling. Thats simple yet valuable step in any data management / standardization project. I have described its importance, approach and solutions in my blog – http://manageyourdata.blogspot.com

Taxonomy is also very important when you think about Information Architecture for portals, collaboration sites, etc. Without a well defined and well thought out taxonomy, it would result in much chaos – a good taxonomy helps to mitigate potential chaos. I also allude to this in my post around Enterprise 2.0 and Governance (found at http://thefuturevalueofbusiness.com/enterprise-20-and-governance.htm).

Raveen