(61) Himali Kumar - Reoccurring Data (Quality) Audits

We hear and understand that step 1 in developing a #dataorganization is to clean up the data. This often includes data harmonization from multiples sources, classification, mapping, labeling, etc. Most often it usually stops there. We cleaned it... we are done, right? No, not really. In my conversations with Himali Kumar an IT and Data Executive, new data is constantly be added into the enterprise data set. From information provided by new customers, inclusion of new data sources, changes in data sources such as POS system, customer preferences, etc. Question: How do you gather and ensure value in enterprise data if the data is constantly changing? If you are not leveraging a #datagovernance program with resources dedicated to continue to re-evaluate the data, then the organization is already falling behind or creating a negative perception between the company and the customer. Examples of challenges that can occur after the initial #datacleansing include: - duplicate customer data leading to sending the same message multiple times not realizing it was the same person (from a data perspective) - data schema issues leading to many man hours of rework - #dataaccuracy issues causing outages to internal and external stakeholders A decision-based organization is only as good as its data. If the data is constantly changing, then it is critical to continue to evaluate (or audit) the data to minimize gaps and mitigate data strategy challenges. My conversation with Himali, also highlighted: - Impacts of not performing regular reviews (audit) of the data - Three (3) areas critical to developing a data organization - the minimum to get started with - Establishing tolerances / risk thresholds to enterprise data - Who is really the end user? - Achievability of #dataquality