If information lacks quality, then it’s not really information
Thats what Lou Agosta a principal analyst at Forrester Research said in 2005 for an article which still holds true and will hold true for a long time to come. The news article in in Techtarget titled “Information quality market to reach $1 billion” It’s one thing to claim having a great customer and marketing database with hundreds of thousands or even millions of records but the value doesn’t lie in the data or bulk of data gathered itself. The value lies in it’s usefulness and what it helps the company do. Whether its for marketing programs, improving customer care or helping make decisions, the value is how that data translates into into information which is useful to the organization. There are two specific areas which data quality can be broken up in to.
The first type of quality refers to the validity of a data record. For example, a simple record which states Benjamin Davis is IT Director at XYZ company located at ABC address and can be contacted at 123-456-7890 tells you Mr Davis is a prospect responsible for IT and provides contact details which is all useful information in connecting with him. In this case if either his name, job title, location or contact details are no longer correct, that data is no longer useful. It can’t be considered “useful” information. Dead contacts (no longer within the same company or holding the same job role), bad contact data and invalid addresses are all signs of deteriorating data quality which can be avoided with a little more attention.
The second type the less obvious type and doesn’t refer to the validity of the data but rather “does it give you all the information you need?” For example, lets look at the same record where we know Benjamin Davis is IT Director at XYZ company located at ABC address and can be contacted at 123-456-7890. Similarly assume you have 5000 similar prospects and the same data. This is useful to some extent but if your intent is to use this data to execute an email campaign… it falls short. It lacks email addresses so from the usefulness perspective, it lacks quality till its complete. This type of quality is subjective to the purpose of the data and what information is needed of it to make it useful. A company that sells java software testing solutions would perhaps need to know if their target accounts use Java in addition to the account and contact data they already have. A company selling a Salesforce.com dependant application would need to know if their target account uses Salesforce CRM in addition to their standard data. A company that sells services to small local branches of large banking corporations would need to know the locations and network of the branches. True “quality” data is rich enough to help you with all the information you need to know to make better decisions and help drive better results. If your company data is lacking certain data points which would make it more useful, data enrichment could help complete it with what you need to know. Quality is subjective to the purpose of the data and if its not complete, it lacks quality.
If it lacks quality, it isn’t really information.