Highlights of a Data Quality Love Fest – IDQ Summit 2014
Over 100 hardcore data quality professionals gathered at IDQ Summit 2014, held in Richmond, Virginia from 10/6 – 10/9.
The agenda was fully packed. Over 40 sessions were launched during the two-day main conference period. In addition, there were two full-day of tutorials being held before and after the conference, covering topics from data quality for beginner to how to establish lean, agile data governance program. I am happy to report there was no shortage of passionate debates over various subject areas among us data quality maniacs. We “argued” about things like what exactly is “good quality” data; where do ethics stand in data governance practice; should data privacy become fundamental human right. We got an update on data management challenges in EU; shared data quality practices in various industries, and had a glance at data governance basics and operating models.
As a first timer for IDQ Summit, I found it refreshing to meet other data quality professional and hear their stories, it is also reaffirming that many of the practitioners in this space share similar views as we at Informatica has been advocating. The consensus on the importance of quality information will help increase the adoption of data quality tools and drive the development of true data –drive culture in many organizations. Data alone has little value, only the data that meets the quality requirement becomes real asset.
My key takeaways are the following:
- It is well understood that data quality is a business process and involves multiple stakeholders – business unit owns the data and makes requirements about the data they need; analyst prepares and manipulates the data for the business; IT enables an efficient data quality process by recommending and implementing the right tools for analysts and business users.
- Data Governance is a journey and takes in different forms. Operation-wise, it can be centralized, decentralized or hybrid, depending on the culture within a particular organization. Senior management sponsorship is critical in making it a success and sustainable. It is recommended to start the practice within business unit rather than IT, as business often has the ownership of the data and understand the context of the data.
- Quality of the data is not an absolute measure – it is largely agreed that quality of the data is only considered “good” when it meets the right requirement at the right time to the right people. When the requirement is no longer valid, or the person responsible for the data has moved away, or the timing has changed, then ”good” data is no longer good.
- Data quality means different things to different people (boy have we heard that before!). Therefore, it is suggested that persona should come to consideration when discussing data quality with people in different roles.
- Should ethics be considered in data governance practice? Some stated ethics could be different from one country to another, however, aligning our personal code of conduct with the professional code of conduct is a good thing. When in doubt, ask yourself: “Would I do this to my loved ones? If not, then why should I do this to others?” Loosely quoted from one of the speakers. Well said.
- Collaboration between IT and business is critical to the success of data quality process. Tools that help facilitate this joint effort are needed to enable a true data-driven culture.
- Metadata should be considered a key component while implementing data governance practice. I like the what was presented by Ron Klein, a seasoned metadata practitioner from KPMG, which says: “Metadata provides a pedigree to the information: what the information is, where it came from and how it got there, what systems use it, its relationships to other information”. Metadata is important, period.
I welcome your views on those topics and would love to hear your stories. Meanwhile, I invite you to visit our Linkedin group discussions on data quality and related topics here.