Practical Guide to Managing Project Risks brought by Data Quality Issues
Building on the previous article “Data Quality Issues” and the “Iceberg” analogy from a natural perspective, we will delve further into this subject by examining the inherent characteristics of data.
Typical Issues and Solutions
The 5 Vs of Big Data refer to five crucial dimensions, which is Volume, Velocity, Variety, Veracity and Value, characterize big data challenges and opportunities. Categorizing typical data quality issues in the context of the 5 Vs of Big Data can help in systematically understanding and addressing these challenges. Below is a snapshot that outlines this categorization with descriptions and examples of typical issues for each dimension, and possbile solutions:
Soluion in a Time-bounded Project
In the lifecycle of a project, the risks brought by data quality issues are not completely avoidable , so how can we manage them in a time-constrained initiative? There are 2 useful practise that we would adopt in digital tranformation initiatives:
- Early identification and assessment in the project planning phase
- Involve data experts early in the project to assess the quality of existing data, and perform initial data exploration to identify potential quality issues before they impact the project
- Conlude the exploration results into pre-conditions of the project plan and actions to be taken by relevant stakeholders. Here’s a template for tracking and managing data quality issues in a project context. Each row represents an individual issue, and the columns provide a structured way to document and address it
- Adopt an agile methodology that allows for regularly review data quality reports and issue list to catch and address issues promptly.
- Establish key data quality metrics and benchmarks, and create a detailed data quality assurance plan that outlines processes for checking and improving data quality at various stages of the project.
2. Contingency Time and Resources in the project excution phase
- Develop a risk management plan that includes data quality risks, and keep stakeholders informed about data quality issues and their impact on the project. Below is a RAID risk management template that you can use to document and keep track of risks specifically:
- Build extra time and resources into the project plan to address unexpected data quality issues.
By setting these pre-conditions and corresponding actions, the project plan becomes more robust, with a strong foundation built on accurate, relevant, and well-governed data. This approach minimizes surprises and enables the team to address potential issues proactively, thus increasing the likelihood of project success.
