Data Quality – who’s responsible?
The answer is – everyone in the business. Unfortunately, all too often no one seems to take responsibility or realise its value.
So just why is data quality so undervalued? The underlying reason is that there is a denial mindset regarding ownership of the data content. In many organisations, IT builds and owns the database container and the data users take ownership of the content, or perhaps not. It’s the classic ‘Everybody, Somebody, Anybody, Nobody’ no ownership story:
Data quality management is an important job and Everybody is sure that Somebody will do it. Anybody can do it, but Nobody does. Somebody got angry about that because it was Everybody’s job. Everybody thinks that Anybody can do it, but Nobody realises that Everybody won’t do it. In the end, Everybody blames Somebody when Nobody does what Anybody could do.
Accurate data is undoubtedly the cornerstone of industry, but a lack of standardised data prevents efficient information exchange between departments and subsidiaries and impedes decision-making and understanding of business problems. Paradoxically, many businesses have masses of data and little information, whilst their employees have masses of information and very little data:
The information paradox
How do businesses manage data quality?
- Ignore it—bury their heads in the sand
- Accept it—like global warming, litter, poor grammar, death and taxes
- Expect someone else to take responsibility, like IT, or get temps in to fix it
- Treat the effects, rather than deal with the root cause
- Take purposeful action to solve the problem
Why does it matter?
This can be summed up in one word – profit.
Businesses squander millions annually by relying on inaccurate or poor quality data, which causes unnecessary waste and impedes decision-making and understanding of business problems.
Any data not fit for purpose which is used by the business means scrap and rework and bad decisions which in turn cause waste. Waste translates to loss of bottom line profit.
Improving data quality:
- Improves profits
- Reduces the costs involved in undeliverable and duplicate mailings
- Increases customer satisfaction
- Improves brand image
- Delivers insight and a single customer view
- Ensures accurate business decision-making
How does poor quality data show up?
Data has no value unless it can be used for confident decision-making and forecasting, accurate, targeted mailings and successful marketing campaigns. If an organisation suffers from any of the following symptoms, it has a data quality problem:
- Poor response to direct mail campaigns, caused by incorrect, inaccurate or incomplete addresses
- Lack of a single customer view, caused by duplicate, inaccurate or incomplete data
- Inability to link/match disparate databases, caused by incompatible data formats
- Increased marketing costs, caused by undeliverable mail, postal returns and duplicate mailings
- Inaccurate forecasting and business decisions, caused by incomplete and inaccurate data
What is the impact?
Data quality impacts on all areas of the business:
- Reduced profits
- Increased costs
- Reduced customer satisfaction
- Brand damage
- No single customer view
- Making bad decisions with a high degree of certainty
- Business intelligence not delivered
- Data warehousing doesn’t provide data quality
- CRM/ERP systems don’t deliver
- Cost of duplicate mailings, data scrap and rework
Who should take responsibility for data quality?
Everyone – poor data quality significantly impacts your bottom line, so it’s a business problem, not exclusively a problem for IT, marketing or any other user. By taking control of data quality, companies have a real opportunity to reduce costs, increase efficiency and dramatically improve their market positioning.
The impact on the organisation, from the improvement to the bottom line through eliminating postal waste and data scrap and rework, through to making better-informed decisions based on accurate data, better customer care, lifetime value and improved brand image, is substantial.
Where do you start?
Get out of denial and own up to the fact there is a data quality problem; get executive buy in.
Assess where the greatest damage is being done to identify where fixing the problems will generate the greatest return. Unless you have unlimited time and money to spend on a data quality project, think 80:20 – break the problem down into its critical parts and sequentially fix the 20% of problems which have 80% of the financial impact.
- Look at what can be achieved tactically within each data source to improve data quality
- Identify the defective business processes which often create a cycle of “correction and corruption”
- Research data quality improvement tools which will help solve the critical problems identified
- Work diligently to first identify, then correct and prevent the unhealthy data defects which damage the effectiveness of data systems deployed in your organisation
- Keep doing it!