The phrase “data rich and information poor” was first used in 1983’s bestselling business book In Search of Excellence, and it’s more relevant today than ever before. In our technology-led world, businesses collect huge amounts of data every day. Knowing what to do with this data can be difficult - bad data quality can lead to inaccurate and slow decision-making.
Your organisation could be sitting on a data goldmine – you might have a patchwork of operational systems and applications with different data which could be used to improve decision making and uncover insights. However, it can be difficult to understand exactly how to synchronise all your data to allow this to happen.
Transforming your goldmine of data into actionable and valuable information is a skill, but one your employees are more than capable of achieving. Learning how to get the most from your data is essential before you can use it, and this is where the six dimensions of data quality come in.
Widely recognised as a key indicator of data quality, the six dimensions of data quality make it easy to understand the quality of your data and find areas for improvement. The UK Government recommend using these six dimensions, as defined by the Data Management Association UK, to determine the quality of your data:
Let’s look at each in turn.
Dimension One: Completeness
Completeness refers to how comprehensive the data is, and if it meets the expectations it was collected for. Completeness of data does not mean having an infinite amount - it’s about whether the data you have meets the specific purpose it was collected for. For example, if you require a customer’s first and second names and some include a middle name and some don’t, the data is still complete if you do not have every middle name. To test the completeness of data ask yourself:
1. Is all the necessary information available?
2. Are any of the values missing key information or elements?
Dimension Two: Consistency
Consistency means data across all systems are in sync and the same across the whole organisation. For example, are offices which are closed still being included on sales reports or former employees still appearing on the payroll? To test the consistency of your data ask yourself:
1. Are data values the same across all sets?
2. Can you see any clear conflicting data within your system?
Dimension Three: Conformity
Conformity refers to your data all fitting the same set of definitions such as size and format. For example, you may want all timestamps to be in 24-hour or 12-hour format, but it should all conform. To test the conformity of your data ask yourself:
1. Do all data values fit within the formats specified?
2. Does all data comply with the formats specified?
Dimension Four: Accuracy
Accuracy is vitally important - the data should correctly reflect the thing it is describing. For example, having the correct address for customers in your data base. The accuracy of your data can impact operations so must be verified and checked regularly. To test the conformity of your data ask yourself:
1. Do you regularly check for updated information regarding customer data?
2. Are there incorrect spellings within your data?
3. Does your data accurately match the real-world values they represent such as addresses and names?
Dimension Five: Integrity
Integrity is the validity of data across connections and relationships, and tests whether your database allows for connections between data to be made. For example, a customer database should have all their information stored together and connected, with no names or addresses without their counterpart. Not linking related records together makes for a highly disorganised and potentially unusable database. Ask yourself:
1. Are all records in your database complete and relationships properly connected?
Dimension Six: Timeliness
Timeliness can’t be ignored as it ensures the data you hold is relevant and up-to-date. Timeliness of data is important for quarterly or annual reports and to remain in line with GDPR guidelines which state certain types of data must be stored for the shortest time possible. It is important to regularly check and ensure you only keep the data necessary for your requirements and dispose of unneeded data securely.
The six dimensions of data quality help to ensure the data in your business is fit for purpose and your workforce can make the most of it, turning it into a valuable resource to drive decisions.
The right data visualised and explained in the right way can drive decision making within the business. Data storytelling is a highly effective tool for getting the most from your data.
Data storytelling allows you to turn raw numbers into something meaningful. There are many creative ways to display data which makes it easy to see its purpose, show patterns and help drive better decision making.
There are some great examples of data storytelling out there such as Why Do Cats and Dogs…by Nadieh Bremer which uses Google Trends search data and stylish visuals to show the comparative behaviour of cats and dogs. For a more serious example, consider The Bill and Melinda Gates Foundation’s Examining Inequality which utilises the latest data to show how geographical location and gender influence the opportunity for a healthy and successful life. Transforming data into stories and visuals requires certain skills. Like everything, these skills can be taught and trained to willing employees.
Your business needs data professionals and employees who have experience and understanding of data analytics. You need people who can confidently utilise your data and ensure it isn’t left unmined and useless to the business. Hiring data science experts is hugely valuable but comes at a great cost and even large corporates are limited to a relatively small data team. Taking the lead in driving data skills development across your organisation is fast becoming a critical focus for Talent and L&D teams.
The World Economic Forum’s Future for Jobs Report 2020 found 73% of organisations will provide reskilling or upskilling opportunities for their employees by 2025. A lot of reskilling will be in data analytics, as it’s become so integral to business operations. A data-led workplace culture ensures all employees have the basic prerequisite skills to make the most of all the data they handle. Of course, data analysts and scientists have the skillset to make the most of data and transform it into valuable information. However, a good knowledge of the value and importance of data should be standardised across the workplace.
Building up mountains of unused and untapped data is a waste of your organisation’s resources. Taking the time to understand, analyse and transform this data into tangible information is essential to growing any business in our data-driven world, and a priority for learning and development teams.