South Africa’s graduate unemployment rate reached 12,4% in the second quarter of 2026, highlighting the difficult transition many young people face from tertiary education into the workplace.
By Nancy Madiba, head of department: information systems and databases at Richfield
While graduates continue to have better employment prospects than people with lower levels of education, the figures raise an important question: are graduates being equipped with the capabilities they need to translate what they have learned into workplace value?
For employers, this is not an abstract concern. According to the 2024 IITPSA ICT Skills Survey, 65% of employers say skills gaps have a moderate to high impact on their businesses, with particularly acute shortages in AI and machine learning, data science, cybersecurity, and business intelligence. Employers also ranked professional experience ahead of graduate degrees when assessing candidates, with many reporting concerns about graduates being insufficiently prepared for the workplace.
This is why data literacy is becoming a core graduate attribute, rather than simply an IT skill.
Data literacy does not mean turning every graduate into a data scientist. It means equipping students with the skills and attributes to understand information, assess it critically, identify what matters, and use it to inform decisions. As data becomes embedded in everyday business processes, these capabilities are becoming relevant across a variety of disciplines.
Graduates entering the workplace will encounter data in almost every function, and they need to be able to interrogate information, recognise patterns and inaccuracies, and apply what they are seeing to the particular problem in front of them.
For employers, this changes what ‘job readiness’ should look like. A graduate who can use a spreadsheet is useful. A graduate who understands what the data is telling them, knows which questions to ask and can use that information to solve a business problem is far more valuable.
The same principle applies to AI. Knowing how to prompt a generative AI platform is not enough. Graduates need the critical thinking to evaluate the output, challenge it, fact check it and use it appropriately. Otherwise, the technology risks becoming a substitute for understanding rather than a tool that strengthens it.
This has implications for tertiary educations institutions, too. Data literacy should not be an optional or stand-along module. It must be embedded alongside academic fundamentals, with students given opportunities to apply technology to real problems and develop the judgement to use it effectively.
It also means keeping curricula current. Industry input can help institutions understand which capabilities are emerging, while lecturers themselves need to continuously develop their knowledge. The technology students encounter when they graduate may look very different from what was being used when their qualification was designed.
For employers, this should encourage a slightly different approach to assessing graduate candidates. Rather than looking only at the qualification on a CV, it is worth asking what the graduate can do with that knowledge. Can they analyse information? Can they solve an unfamiliar problem? Can they use technology without becoming dependent on it? Can they explain the reasoning behind a decision?
The real measure of a future-ready graduate, then, is whether they can combine what they know with the data and technology available to them, and use both to make better decisions.