At the 2025 ACSG conference, Dr Angela Marsburg and TTS CEO, Prof Hennie Kriek, delivered a compelling presentation on how AI-powered passive data can redefine how organizations understand and manage workplace well-being. In their presentation (and the following article) they introduce one of TTS’s latest product partners, Welliba, who have designed an innovative, real-time platform that measures employee sentiment and experience using the power of Artificial Intelligence (AI).
Why workplace well-being matters
The session opened by reaffirming the empirical link between employee well-being and productivity, with findings showing happier employees are up to 20% more productive and 37% more effective in disciplines like sales. Yet despite rising awareness, data from the Gallup 2024 Global Workplace Report for Sub-Saharan Africa trends show that:
- 74% of employees are struggling, and
- 75% are watching for or seeking new jobs
This disconnect reflects a growing need to rethink how well-being is assessed and acted upon, especially given that traditional surveys fail to capture real-time nuances or objective data.
Defining the constructs: Engagement, experience, and well-being
To bring clarity to common misconceptions, the team distinguished between employee well-being, experience, and engagement:
- Well-being refers to subjective feelings of positivity and personal resilience.
- Experience is the longitudinal and contextual perception of the workplace.
- Engagement reflects the motivational and cognitive investment in one’s work.
Each construct is important but well-being is most vulnerable to external stressors and most neglected in typical survey approaches.
The new paradigm: AI-Powered insights
The Welliba approach presents a holistic and non-intrusive alternative to typical employee-based surveys or once-off diagnostics.
By using AI algorithms to analyze passive digital data streams (e.g. communication patterns, social media, third-party websites, etc), the solution enables organizations to:
- Monitor well-being in real-time, without overburdening employees
- Integrate external and internal data sources to improve accuracy
- Bypass or mitigate survey fatigue and low participation rates
- Ensure transparency and interpretability of AI models
Case study: South African applications
A local application of the solution showed how South African data diverges from global norms:
- South African engagement is higher than the global average
- However, “thriving” scores are lower, and intent to leave is significantly higher
These insights illustrate how region-specific passive data can uncover trends that surveys may overlook and supports the implementation of more nuanced, responsive talent strategies by local organizations.
Practical recommendations
To conclude, the presenters offered actionable guidance:
- Prioritise well-being without increasing cognitive load on employees
- Move away from backward-looking approaches and toward continuous insight
- Use AI to identify real-time drivers and blockers of well-being
- Create organisational cultures that are responsive, flexible, and informed by multiple sources of data