Published on the 08/06/2023 | Written by Heather Wright

Organisational culture, people and processes…

Avoiding the high costs of data quality issues and delivering sustainable value to businesses through data, isn’t as hard as some might think.

That’s the view at least of Gartner’s Jason Medd, a director analyst with the company.

Medd is open that data quality issues are costly. Gartner has previously outlined the significant financial impacts of poor data quality, estimating in 2021 that poor data quality cost organisations an average of $12.9 million. And it’s not just immediate revenue impacts, with Gartner noting longer term issues of increased complexity of data ecosystems and poorer decision making.

“Not all data is equally important.”

Roll on two years, and Medd says organisations without an impactful and supportive data quality program in place will face a multitude of complications and lost opportunities. But, he says, data quality isn’t hard to fix and it doesn’t have to take a lot of time.

“One of the mistakes chief data and analytics officers make is taking a technology-centric approach to data quality improvement, with little focus on organisational culture, people and processes to streamline remedial actions,” Medd says.

In today’s data driven world, data quality can be crucial component in the success – or otherwise – of businesses, something highlighted by the frequent view that data is an organisation’s most valuable data – the ‘new oil’. But as the old adage goes, garbage in, garbage out.

And it’s not just the danger of flawed decision making. A recent report from the CSIRO’s National Artificial Intelligence Centre found that data quality is a key challenge impacting AI adoption, with 59 percent of decision-makers polled reporting they were concerned about using AI to analyse customer insights due to poor data quality. That puts data quality up there with security and privacy concerns as the leading challenges for Australian organisations’ adoption of AI.

Medd says analytics leaders need to take pragmatic and targeted actions to improve enterprise data quality, and says there are 12 ‘quick’ actions which can help organisations not only achieve data quality goals, but sustain them for long-term value.

Those 12 steps fall into four categories, starting with focusing on the right things to set strong foundations.

“Not all data is equally important,” Medd notes. “CDAOs must focus on the data that has the most influence on business outcomes, understand the key performance indicators and key risk indicators and build a business case.”

With a business case in place, it’s time to define data quality with stakeholders and establish data quality standards.

The second category is around defining who is responsible and assigning data quality accountability. This, Gartner says, is done through including data quality in data and analytics (D&A) governance initiatives and getting sponsorship from the D&A governance committee; and establishing data quality skills and responsibilities in data stewards from business units and the central D&A team.

Those stewards can potentially look at building real-time data validations where needed to help bridge gaps, along with exploring new avenues for improvement and ‘proactively shifting gears based on priority’.

Forming special interest groups who can benefit from data quality improvement, communicate the benefits and share best practices with business units, rounds out the data quality accountability work.

Establishing fit for purpose data quality and taking action to achieve it are the next steps, Medd says.

It’s a category that requires performing regular data profiling and monitoring, building improvement plans and transitioning to a trust-based governance model.

“To improve data quality it is important to perform data profiling and data monitoring to understand and validate current data gaps and challenges, monitor and build improvement plans.

“Then, CDAOs need to transition to a governance model based on trust to drive enterprise-wise adoption of data quality initiatives.”

Using technologies to reduce manual efforts and get faster results, identifying frequent data quality issues and incorporating the solutions into business workflow, can help businesses integrate data quality into corporate culture more effectively.

“CDAOs should also improve data literacy across the business by installing a data quality culture and facilitating knowledge sharing and collaboration among all the stakeholders of the program.”