Building foundations for good data governance

Government remains plagued by a siloed approach to data governance, a technology conference has heard.

Saul Judah

Gartner VP Analyst Sauh Judah says governance is often confused with compliance. Rather, good governance should be viewed as an enabler that allows an organisation to move forward.

“We need to have solid foundations that allow us to move forward on solid ground,” he told Gartner’s Data and Analytics Summit in Sydney on Monday.

“If we’re going to be crossing a bridge, we need to know that the bridge that we’re crossing is going to hold our weight.

“We’re looking at departments and areas of government that are not asking cohesively what the organisation wants to achieve.

Seven must-have foundations for modern data and analytics governance:

  • Accountability and decision rights
  • Collaboration and culture
  • Trust
  • Value and outcomes
  • Education and training
  • Transparency and ethics
  • Risk and security

Trust-based governance is fundamental, he said.

When it comes to making decisions, accountability is important, and  there needs to be an awareness of who is making the decisions.

“It’s important for us to understand not just where decisions are made but how they’re made,” Mr Judah said.

He says it’s also important to recognise that data analytics comes from all sorts of places.

“To the degree that we understand and trust that information, the trust network or the trust model will help us to land the right decisions through governance,” he said.

When it comes to transparency, it’s necessary to ensure that decisions are justifiable and that organisations can prove where money is being spent.

“Transparency and ethics has been the cornerstone of good governance in the governance realm … for years and years and years, unfortunately with data analytics we’ve not woken up to that,” Mr Judah said.

“To me, the way you handle risk and security is a really interesting leading indicator of how mature you are in your organisation.

“Don’t lose sight of what you’re doing, and don’t lose sight of why you’re doing it.”

Four key questions for data and analytics governance:

  1. Do we need to know what is happening?
  2. Is it safe?
  3. Is it trustworthy?
  4. Can it be leveraged?

Data Stewards

The confence also heard that appointing a data steward on the team can help ensure quality data.

Good quality data should be a cornerstone for every organisation, Gartner’s Senior Director Analysist Melody Chien, told delegates.

Melody Chien

This means taking a proactive approach to ensure that data collected from the beginning is good.

“Whether you want to build a business successfully or you want to be more competitive, data quality is always the cornerstone for anything you want to do,” she said.

“Every organisation needs data quality. There is no shortcut on that. It’s something you have to do. Data quality will not come automatically.

“Poor data will lead to poor decision making.”

Lack of ownership and collaboration can affect data quality and this is a cultural problem that needs to change, Ms Chien said during her presentation at Gartner’s Data and Analytics Summit.

“We always say that the data quality is everybody’s job, that everybody has a responsibility to keep the data right.

“But let me tell you, when you say that it’s everybody’s job, the way it turns out is that it’s nobody’s job.”

The problem can be overcome by assign the role to a specific “data steward” on the team.

They don’t necessarily need to have a formal title. They can be part of the business team or add the responsibility to their existing work.

“It’s very clear in an organisation who is responsible for marketing, who is responsible for sales, but it’s not so clear who is responsible for the data quality,” Ms Chien said.

“The success of your data quality program is dependent on this role – data steward,” Ms Chien said.

Data validation is also important to data quality, and should be completed at the beginning to save cost and time, according to Ms Chien.

“It might take you one dollar to validate the data versus it takes you 100 bucks to fix that data or eventually one million dollars if you don’t do anything at all,” she said.

Data integrity will also help organisations “stay out of jail”, Ms Chien added.

“We want to make sure we’re compliant with standardisation to keep us out of jail,” she said.

Comment below to have your say on this story.

If you have a news story or tip-off, get in touch at  

Sign up to the Government News newsletter

Leave a comment:

Your email address will not be published. All fields are required