The quality of data that data analytics systems use is a core criterion to derive value from such systems for businesses in South Africa, with many private-sector organisations placing advertisements for data quality specialists.

The practical elements of decision-making processes often centre on whether the decision-maker has the relevant data to make the right decisions, says advisory firm KPMG South Africa data and analytics leader Frank Rizzo.

Data analytics systems can provide significant value in decision-making and optimisation, as the systems can compare different sets of data to enable users to identify specific portions of the business that are performing suboptimally and make detailed, complex and informed decisions to improve the company or division.

“Companies are starting to realise the necessity of analytics, but data quality needs to be in place first. Much experimentation is still being done in the South African marketplace on the real-time analysis of data, but there have been no significant successes yet.

“There is a gap in the market to gain insight from the tools that companies are using, which may be caused by a resources issue. Companies need people who understand mathematics, science, statistics, technology and business.”

Rather than only focusing on the volume, variety and velocity of data, the integrity and value of data should also be prioritised, which may be more important to enable robust business decision-making than only the volume of data analysed. However, all of these elements need to be used in conjunction to help decision- makers respond to a specific business problem they need to solve, he notes.

Companies aiming to use data analytics to identify problem areas in the business or to provide a factual basis to inform business decisions must remain cognisant of the quality of data used in these analyses, emphasises Rizzo.

“While knowing what questions to ask and how to ask them demand specific skills sets and understanding of the analytics tools, the quality of data used in the analyses is key to improving the quality of analysis and, consequently, business decisions.”

The volume of data available, as well as the volume of real-time data being produced, is adding to the problem of data quality, but data filtering and cleaning methods to ensure that the data can be used by analytics systems is increasingly necessary, hence, the need for data scientists and professionals.

“Just as business and technology have merged in recent years, so too will the science of data become part of the workplace. There will be significant demand for skilled people in these areas. Data proliferation will increase exponentially and demand for specialists who can filter and extract value from data will also increase,” concludes Rizzo.