In the face of tough competition, marketers face the dual challenges of acquiring profitable new customers and reducing the customer churn rate.
As most marketers agree, it is more costly to acquire new customers than to retain existing ones, so it has become critical for companies to maintain their existing customer base. This means the focus is shifting to relationship-building initiatives, such as customer clubs and other retention and loyalty programmes. A robust customer database is key to running effective programmes of this nature, so marketers and agencies are rushing to collect customer and prospect data in a myriad of ways.
While most companies are aggressive in their approach to data collection, this is often at the expense of consistent and well-planned strategies for analysing and converting the data collected to provide meaningful information.
The focus tends to be on accumulating as much data as possible without really developing strategies to digest the data and use it effectively.
In addition, most organisations do not have a well-defined policy for data cleaning, trimming and junking. Often, even when they do perform such tasks, they're not regular and consistent. In this way, databases become 'overweight', which is as unhealthy for them as it is for humans.
What happens when we are overweight? It is more difficult to manage our body, we tend to slow down, and we become more vulnerable to other diseases, such as heart problems, blockages etc. This has a clear parallel in database health where excess data causes a range of problems like slower processing of data, data errors, and crashing of data files. Such unhealthy databases are a management headache and start losing money rather than making it.
What can we do to avoid this situation?
Adequate planning before any data collection exercise is critical. First it is necessary to identify what data is actually needed and to avoid the temptation to collect unnecessary information. It is also important to specify in the early stages what mandatory/optional fields are required and to finalise data codes and data residence issues up-front so that there is no confusion when it comes to data entry, management and usage.
Secondly, newly-collected data should be checked against existing data for duplication and records modified accordingly. Moreover, regular data cleansing, de-duplication and redundancy checks have to be done to ensure that unnecessary, junk data is trimmed in time. In short, data health policies should be based on the same principles as those advocated for a healthy lifestyle for people:
- Consume only what you need and what you can digest.
- Keep fit and well by exercising on a regular basis.
