3 Reasons People Analytics Stalls, and How AI Fills the Competency Data Gap

A Compelling Vision for HR, Still Far from Reality

People analytics and data-driven HR have been among the defining themes in HR for more than a decade. People analytics is a data-driven approach that comprehensively collects and analyzes HR data to improve the quality of people decisions across hiring, development, evaluation, and beyond. The vision drew attention from companies everywhere: objectively identify employees’ performance and potential, quantify inefficiencies in how the organization runs, and make strategic decisions tied directly to business results.

Yet even today, it is hard to find companies that have successfully made people analytics a core engine of their HR operations. Despite heavy investment and many attempts, most projects ended at the pilot stage or stopped at a dashboard showing a handful of metrics. Why has the vision proved so hard to realize at most companies?

Below, we look at the main obstacles companies have run into while pursuing data-driven HR, then show how AI addresses the central problem in people analytics, the lack of “competency data,” and offers a concrete path to making data-driven HR real.

‍

1. Three Factors That Made People Analytics Hard to Adopt

Many companies tried to adopt people analytics, ran into similar difficulties, and never saw tangible results. The problem went beyond adopting new technology. It stemmed from the inherent nature of HR data and analysis.

(1) Low Trust in Scattered Data (Data Silos & Integrity)

The first obstacle is fragmented data. A model to predict attrition among key talent, a project many global tech companies attempt, is a classic example. Building it requires a wide range of data on people who have left: past performance reviews, compensation history, promotion records, training completions, and engagement survey results. In reality, that data is scattered across different systems. Performance data lives in the performance management system, compensation history in payroll, and survey results in a separate external tool.

Before modeling can even begin, analytics teams often spend months matching employee IDs that differ from system to system, converting performance rating scales that have changed several times into today’s standard, and turning exit interview notes, which exist only as unstructured text, into something they can analyze. Without a single source of truth, it is hard to trust the analysis, and preparation alone consumes enormous resources.

"One of the biggest struggles HR teams shared at the event was the complexity of managing data spread across multiple systems like SAP Success Factors, Workday, Oracle, iCIMS, PowerBI and many more. This fragmentation hinders their ability to pull all people data together and make timely decisions." (Gartner, "Reimagine HR Conference", 2024)

(2) The Expertise Gap Between HR and Data Science

HR professionals understand organizations and people deeply but often lack expertise in statistical modeling and analytics tools. Data scientists, on the other hand, have strong technical skills but may not grasp the complex, subtle context of HR, so they can easily produce analysis that misses the actual problem.

Say a data scientist reports “a strong positive correlation between training hours and performance ratings.” Without HR context, that finding can easily be read as a simple causal claim: “invest more in training, and performance will improve.”

An HR professional, however, might raise a different hypothesis: “Isn’t it possible that employees who were already proactive and highly motivated simply score high on both training participation and performance?” In other words, you need a perspective that questions and tests whether training really causes performance or whether the link is a statistical illusion created by a third variable such as conscientiousness. Without HR’s deep context and critical eye, data analysis can lead straight to the wrong decisions.

(3) The Question of What to Measure

The heart of data analysis is measuring what matters, and measuring it correctly. But the competencies and skills that most shape organizational performance, such as leadership, collaboration, creativity, and contribution to culture, are inherently very hard to quantify. They show up in specific behaviors and outcomes, and traditional HR systems had no way to capture them as data.

So companies had to rely on proxy metrics that are easy to measure, such as training hours, tenure, and performance ratings, which show only a fragment of the picture and never explain what is really going on. Once teams end up managing only what they can measure, the analysis has little real influence on decisions. This became one of the biggest headaches in pursuing people analytics.

‍

2. New Possibilities for People Analytics in the AX Era

These challenges slowed the progress of data-driven HR. But generative AI and large language models (LLMs) offer a new paradigm for solving them and are opening a genuine era of AI transformation (AX).

Rather than aiming to be a people analytics service that integrates and analyzes every kind of HR data, from culture and attendance to performance and turnover, Telta focuses on the most fundamental data, the data that decides whether all of that analysis succeeds: competencies and skills. By providing a solid foundation of trustworthy competency data, Telta acts as a key enabler that raises the quality and reliability of people analytics as a whole.

(1) AI-Powered Skill Set Building and Customization

Companies used to hire consulting firms to run competency modeling projects that took months. Beyond the heavy cost and time, the resulting competency frameworks were often so disconnected from reality that they were quietly shelved.

Telta solves this problem with AI. It proposes custom skill sets suited to each industry and role based on global job data, then trains on the company’s internal data, such as job descriptions and business reports, to build a tailored skill set that fits the organization’s unique context and strategy in a short time. It is the most practical and efficient way to solve the problem of what can’t be measured.

(2) Automating Competency Assessment

Reliable data requires ongoing, repeated assessment rather than a one-off exercise. Telta automates the entire competency assessment process, dramatically reducing the operational burden on both HR and employees while speeding up data accumulation. In one case, a client that ran its assessments through Telta cut the operating workload of the team responsible by about 70%.

AI-powered automation frees HR teams from the endless work of competency modeling and assessment administration so they can focus on higher-value work such as developing people and improving culture. It efficiently develops assessment items from the skill set, scores and analyzes responses in real time, and automatically generates a draft of an in-depth report covering each person’s strengths and weaknesses and the competency profile of the organization as a whole.

(3) Making Organizational Competency Visible to Support Strategic Decisions

The structured competency data generated by Telta’s AI-powered solution makes an organization’s previously invisible competency assets visible. HR and leadership can now answer critical questions with data: Which skills does our company have, and how much of each? How wide is the gap between the competencies our future strategy requires and where we are today (the skill gap)? Who has the best combination of competencies for a new project? Companies are already turning this data into tangible results, for example by using the competency data from the solution to measure the impact of training aimed at strengthening core competencies.

‍

‍It’s Time for HR to Restart Data-Driven Innovation

The vision of data-driven HR was never wrong. What was missing was the reliable data needed to make it real. In particular, objectively measuring and capturing as data the competencies of employees, an organization’s most important asset, was a hurdle existing methods struggled to clear.

AI is the engine that can overcome these obstacles and deliver the value data-driven HR promised. Data-driven HR that hires and develops talent based on objective data, manages potential organizational risks before they surface, and unlocks the full potential of every employee is now within reach.

It’s time to start HR’s next chapter with Telta.

‍

Related Reading

• What Is People Analytics? Why HR Data Matters More in the AI Era
• AI-Era Talent Strategy: Filling the Skills Data Gap in Your HCM and HRIS

‍