Data is crucial for every enterprise to take the right decisions that can help business flourish. For example, the number of transactions happening in a particular ATM (Automated Teller Machine) over a period of time can tell whether the ATM is being utilised effectively or can be removed. Similarly, if we take a retail chain, data collected over a period of time will help in understanding demand for different products at different times, and ensure that the stock is available during those times.
The data collected can also facilitate opening another retail chain according to the demand in a particular area. The reservation data collected by the Railways over a few years can be useful in providing better service, especially during festive seasons. These are simple examples that emphasise the role data plays in building business intelligence.
Need for quality
Considering its importance, it’s imperative that data be trustworthy. ‘Data quality’ is a process that helps ensure this and has become indispensable in today’s IT space. It helps in straightening out simple anomalies in data, for example, ensuring that wherever a date of birth is mentioned, the format is the same.
For instance, a person’s birthday could be misinterpreted as 06th May 1980 instead of 05th June 1980 depending on how the system interprets 06-05-1980 (as MM-DD-YYYY or DD-MM-YYYY). Data quality can also deal with other aspects of data standardisation such as ensuring that all variations of an abbreviation like US, USA, U.S, U.S.A are stored as USA.
There are more advanced processes that ensure validation of the addresses maintained in a system. For example, for a country like the U.S., given an address, the data quality process can help find out whether the address is valid or fit for delivery, depending on various factors such as the state, city, its corresponding ZIP code, etc.
Other advanced variants of data quality use involve concepts such as ‘house holding’. For example, a bank maintaining customer information can use the concept of house holding to group records belonging to the same home/address (or household). If the bank sees them as individual customers, their worth could be less. But with the concept of house holding, the bank can consolidate records and identify how many accounts come from the same household or family. This could help the bank in serving its customers better because, as a household, this family can bring in a lot of business.
Similarly, an insurance company that needs to send correspondence for premium payment can post one for the entire family if it can consolidate data based on address.
This can definitely bring in significant savings in postal costs. Some insurance companies like to wish their customers on their birthdays. However, a person could end up getting three SMSs if he/she holds three policies. If the company consolidates the policies based on the policy holder, it can enhance customer experience to a great extent.
There are several business needs that make data quality an integral part of every data-driven enterprise. There are many software vendors in the market who provide out-of-the-box solutions to achieve trustworthy data using data quality process.
(The author works for aninformation technologyorganisation)