- Poor data quality is a problem facing most organizations.
- There are a variety of hard and soft costs associated with poor data, including increased direct marketing costs, lost revenue, inadequate financial reporting, unfulfilled orders and damaged customer relationships.
- Despite the costs and headaches associated with having poor data, most organizations struggle to define and implement a strategy for effectively improving data quality.
Our Advice
Critical Insight
- While the business must be accountable for data ownership, IT needs to be proactive in providing solutions for data quality problems.
- There are five types of interrelated data quality issues: data duplication, stale data, incomplete data, invalid data, and data conflicts. Each problem requires a combination of corrective and preventative measures by the business and IT.
- Some methods for solving data quality problems are more effective than others. Mass cleanups and strong data validation are more effective than user-training and vendor-supplied tools.
Impact and Result
- Mass cleanups will solve most data hygiene problems (i.e. duplicate, incomplete and invalid data). Stale data can be fixed through external source integration.
- Assigning data stewards will improve accountability for data accuracy.
- Clean data means improved marketing campaigns, decision making and operations planning; both hard and soft costs are reduced dramatically when corrections are put in place.