How to Clean Up Dirty Data

dirty data

The importance of cleaning up your marketing database can never be understated. Failure to do so may result in total waste of the marketing effort and the associated time and investment.

Cleaning up dirty data can be a nightmare, depending on how much time and effort you have already invested in it. The following methodology can help:

1. Develop a Baseline: The first step towards resolution of a problem is to understand the problem. Marketers need to have a clear understanding of the existing stage of the data and the extent to which data drives business. The quality of the available data may be gauged by determining the extent of missing or false information and duplicate data, and identifying sources from where errors creep up.

2. Develop an action plan to clean the dirty data and ensure that the data maintains a determined quality level. This may require involvement of the IT, sales and other teams apart from the marketing ream. A “marketing steward” with specific responsibility for data quality would improve coordination. The way to implementation is to define the required quality goals and establish priorities.

3. Implement the action plan to improve quality. This may include ways such as:

  • Specific plans to reduce errors during the data gathering processes such as implementing drop down lists or predefined numeric ranges wherever possible to enforce consistency, automated form filling, standardized formats and more.
  • Progressive profiling or asking information from consumers in stages as they progress in the lead nurturing stage to avoid prospects filling in something to get over the process at the initial level
  • Validating the information in the database from external sources
  • Prioritizing or white listing sources to collect the data
  • Establishing a regular system to update the database on a periodic basis
  • Ensuring relevancy of data. For instance, the contacts in the billing list are not the people to try and sell the products to. Especially in large organizations, the people footing the bill will be different from the people actually using the product or authorizing its purchase.

Measure the effectiveness of such actions and make the necessary changes. Cleansing marketing data is an ongoing experience and the best approach is to keep on fine tuning the plan based on priorities.

Learn more about big data and data mining here.