Leverage customers' insights to drive Data Quality strategy
The importance of defining and implementing a Data Quality strategy for an organization is well known. Here is an example - The U.S. Attorney General's office stated that approximately $23 billion is lost in fraud or inaccurate billing.[1]
It is necessary to define the different Data quality metrics that need to be measured and monitored. The amount of data being stored in the organizations is very huge. The business processes for capturing this data and transforming it to different forms for different purposes are also very large. It is practically very difficult to define and measure the data quality across all the data flows consistently and continuously.
What strategy would you use to prioritize the metrics?
There is another critical input that can help in identifying the priority data elements that need to be monitored - Listening to customers' voices in social media.
The proliferation of social media has resulted in increasing avenues through which the customers express their opinions about the products and services they consume. The customers talk about the products they like, the issues they face and even provide suggestions for new features for the products. The organizations need to listen to the customers and implement a framework to feed these opinions as one of the key inputs while designing the Data quality strategy program.
What can be mined from customers' Social media conversations?
The key metrics that can be mined from customers' conversations include both the quantitative and qualitative aspects of the conversations in the social media. If there is a sudden increase in the conversation about a product, this implies a stronger buzz about the product. The buzz could either be positive or negative or neutral. Analyzing the sentiment of the conversations, it can be classified whether the buzz indicates a risk that needs to be addressed.
An example scenario could be - a large number of customers of a state like California talking about the overcharge in their bills for a particular month. The Data quality team needs to alert the stakeholders when there is a huge buzz in the social media and the sentiment is negative, while listening for the conversations related to billing.
The conversations can be monitored for the following purposes also
- Reputation management - assessing the risks for the organization, including the management reputation, marketing strategies, financial management, CSR activities.
- Brand monitoring - assessing the risks for the individual brands, products or services of the organization, tracking the buzz generated during product launches.
- Competitive analysis - assessing the risks for the competitor brands, positive buzz generated during competitor product launches
[1] Tara Eck, "Health care companies renew compliance focus", Nashville Business Journal
Comments
Interesting idea that merges unstructured text mining to derive added value. Of course, challenges abound to practically extract value from the social media and conjointly analyze with structured internal data streams to generate meaningful metrics. In the same way as this is applicable to data quality, even management metrics can be extracted in a similar fashion. Has this been implemented anywhere? May we touch base to explore how this concept can be extended? Please ping me at amol.patil@infonitive.com to discuss.
Posted by: Amol Patil | October 26, 2010 3:11 PM
Amol, Thanks for your comment. Here is a case study on how Twitter messages are interpreted and used to mine real-time information on customer experienced problems.
http://www.technologyreview.com/printer_friendly_article.aspx?id=26668
Posted by: A. Sankara Narayanan
|
November 7, 2010 4:58 AM