We give you a quick self-assessment test to help you get started.

Data science and artificial intelligence are constantly in the headlines. We have all heard of self-driving cars and trains, robots that build houses, factories that run themselves, and computers making medical diagnoses. AI has been helping make advancements in numerous scientific fields and industries, including healthcare, finance, education, and transportation among others. 

We hear incessantly that the new programs driven by data-based algorithms are soon to take over the making of business decisions – currently requiring human involvement – and potentially reducing the numbers of jobs. It’s debatable how fast this will happen. 

But how can we be sure that these new automated programs are truly reliable?

The newest techniques can be used –and lots of design and effort put into these new decision making systems. But all artificial intelligence or programmatic systems are dependent on one thing –good quality and reliable data on all facets of the areas related to where they require to make decisions. If they don’t get that –then the output can only be as good as the input –i.e. “rubbish in - rubbish out”.

Also if we look at our organization currently – we can consider it as a similar system making decisions based on the evidential data available in the datawarehouse. 

So how good is the quality of decisions being made? Is the input data good enough to create good output decisions?

A Forester survey of 164 professionals related to Data Governance in US and UK organizations commissioned in 2015 came to the following 4 conclusions:

 

  • Enterprise-scale analytics capabilities require vast and varied data sources
  • Data quality is imperative and a major challenge
  • Organizations that lack a strategic approach to quality fail to keep pace with business needs.
  • Organizations that create a range of data governance structures reap quality and competitive benefits.

How can you tell whether your organization is operating efficiently using correct and quality data?

From where you are in the organization it can be hard to know what data improvement practices are being undertaken. But there are indicators you can look for to give you some level of insight as to the effectiveness. You might not be able to measure all of these, but they can give you an idea as to where your organization currently stands.

It is also important to ensure that data responsibility is taken by all departments of the organization –not just the IT and Business Intelligence areas. Even though these technical areas may be responsible for looking after the data repository technology, they are not the main business owners of the data and may not understand the business meaning of the data and understand its flow throughout the organization.

Here are 10 indicators to rate whether your organization is operating based on good data intelligence – or making decisions based on incorrect or incomplete information. Score yourself on the number of times you answer ‘Yes’ to the questions below:

 

  1. Are you able to easily find data you need to perform your analysis covering what you need in the organization? 
  2. Is there a location where you are able to find data definitions –and are they updated regularly?
  3. Do these data definitions allow you to understand where the data was originally created and how it has been manipulated along the way –in easy to understand business terms?
  4. Is there a way to be able to report data issues e.g. incorrect or missing data?
  5. Are there allocated roles (sometimes called data stewards) to monitor and be responsible to identify data issues from each of the business areas where data is collected or modified?
  6. For these roles described in question 5, - do you know if they are regularly trained and updated –and do they have time to do their role adequately well? 
  7. Are there measures and goals for data correctness and completeness?
  8. For anyone analyzing or making decisions from data, is there regular training available on how and where to find data and the descriptions –and how get started on analyzing the data etc.?
  9. Can you see that data is being updated regularly as expected or documented?
  10. Is there a regular email, newsletter or intranet page updating everyone on new data features or updates- and where to find more information?

Congratulations – you have taken the time and made a self-assessment of your organisation’s Data Governance Effectiveness. 

Score yourself on the number of times you answered ‘Yes’ to the questions above: 

How did you score?

8 or above above 
Excellent –your organisation appears to have Data Governance well in hand. If you want to make your organisation even more competitive than it is now, then there’s always scope to improve efficiency and effectiveness in all areas. One idea is to make data governance meetings more productive and exciting by exploring new opportunities in using data –and potentially new types of data not yet explored.

6–7 
Your organisation is doing well –but could do with a tune up in the few areas it missed out on gaining points above.

4–5
Your organisation could become more productive, increase customer satisfaction, reduce mistakes, create new opportunities and gain advantage over its competitors by addressing many of the above points. 

 <4
Your organisation needs to take immediate action to stay in business –as competitors will be offering a better service to your customers and taking advantage of opportunities your organisation is not even aware of.

What can you do next?

Your in-house Data Governance area should be able to update you on the current initiatives making improvements in the areas where you answered “No” to above.

If you see that some improvements can be made –then it’s worth having a chat with a Data Governance practitioner to discuss your rating above – and to find out what actions should be taken next. These actions should be practical and deliver the most business benefit.

QVARTZ Analytics have experience in performing rigorous data governance assessments quickly and easily –then creating a data Governance strategy defining data strategy goals. Finally, they then create a roadmap showing how to achieve the data strategy goals with small bite-sized projects.