Explaining how data quality improvement will lead to business outcome has always been difficult. The challenge is that there very seldom is a case where you with confidence can say “fix this data and you will earn x money within y days”.
Not that I have not seen such bold statements. However, they very rarely survive a reality check. On the other hand, we all know that data quality problems seriously effect the healthiness of any business.
A reason why the world is not that simple is that there is a long stretch from data quality to business outcome. The stretch goes like this:
- First, data quality must be translated into information quality. Raw data must be put into a business context where the impact of duplicates, incomplete records, inaccurate values and so on is quantified, qualified and related within affected business scenarios.
- Next, the achieved information quality advancements must be actionable in order to cater for better business decisions. Here it is essential to look beyond the purpose of why the data was gathered in the first place and explore how a given piece of information can serve multiple purpose of actions.
- Finally, the decisions must enable positive business outcomes within growth, cost reductions, mitigation of risks and/or time to value. Often these goals are met through multiple chains of bringing data into context, making that information actionable and taking the right decisions based on the achieved and shared knowledge.
Stay tuned – and also look back – on this blog for observations and experiences for proven paths on how to improve data quality leading to positive business outcome.
Liliendahl on Data Quality 