At CUF, people analytics has evolved from a data exercise into a strategic management approach grounded in evidence. The journey began in 2023 with a simple but critical question: once you have the data, what do you actually do with it?
Following the maturity model proposed by Keith McNulty of McKinsey & Company, CUF structured its evolution from “Good Data” — focused on reliability and systematization — to more advanced stages such as “Strong Data,” “Advanced Analytics,” and ultimately “Reliable Predictions.”
As Ricardo Tavares, Head of HR Analytics & Transformation at CUF, notes, these levels are not rigid. An organization may sit at different maturity stages across different areas. The guiding principle? Prioritizing value generation over technical perfection.
Building the foundations: Good Data
The first phase centered on structuring processes and ensuring data consistency. CUF implemented a transactional ERP/HRIS solution and established strict data entry rules to guarantee traceability and coherence.
A telling example is the Full-Time Equivalent (FTE) metric. Given its complexity — including nuances such as service providers and absenteeism — it required several redefinitions before reaching a stable and reliable standard. Automation through Power BI reduced manual errors and consolidated a trustworthy data base for future development.
From accuracy to accessibility: Strong Data
Once data reliability was secured, the focus shifted from “Is the data correct?” to “Who can access it — and how is it being used?”
CUF made a strategic decision to build its HR reporting system internally, leveraging in-house engineering expertise and its existing Power BI-based indicators portal. Two full-time resources were allocated to the project.
The choice offered flexibility and alignment with internal needs, but it also came with risks — particularly delays caused by technical challenges in data access. Another recurring challenge: the temptation to overcomplicate reports in pursuit of perfection, often at the expense of speed.
Still, internal usage metrics indicate strong adoption, laying the groundwork for the next maturity levels.
From analysis to prediction
CUF is now preparing to move toward Advanced Analytics and Reliable Predictions. This phase requires statistical expertise and tools such as R, Python, and SPSS.
Retention analysis is one of the typical use cases: combining structured data (demographics, job characteristics) with qualitative insights (exit interviews) to detect patterns, anticipate turnover, and design more stable and motivating work environments.
However, ambition is balanced with caution. As predictive models and AI gain relevance, ethical and regulatory considerations become central — particularly in light of the European Union’s AI Act and its defined risk levels.
At CUF, people analytics is not treated as a trend or a dashboard exercise. It is an ongoing process of refinement — disciplined, strategic, and human-centered — where data serves not just to describe the past, but to responsibly shape the future.
Related News