A recent study published on Psychology Today analyzes the distortion known as 'measurability bias,' which significantly influences investments and the development of human resources technology (HR tech). This trend leads organizations to prioritize easily quantifiable quantitative metrics at the expense of difficult-to-measure qualitative factors, such as leadership, motivation, and organizational culture. The study emphasizes that the imbalance in resource allocation reflects a deep-seated human desire to eliminate uncertainty, preferring concrete data as a tool for reassurance in high-stakes decisions. Analyzing the structure of industrial/organizational (I/O) psychology reveals a fundamental dichotomy: industrial psychology treats the individual as a unit of measurement, tending toward quantification, while organizational psychology measures groups, opting for qualitative measures. This results in a clear separation between the two fields: the 'I' part (selection, training, performance management, basic HR) is massively funded by technology, while the 'O' part (leadership, culture, motivation) is neglected in favor of technical solutions. [1]
Market data confirms this dramatic imbalance. According to an analysis by 451 Research, less than 2% of the HR tech market sales, valued at $94 billion, are dedicated to qualitative constructs ('O'). Similarly, the HRO Today report for 2025 shows that nearly the entire $3.75 billion invested in HR technologies was allocated to the 'I' section. This massive flow of capital attracts both attention and human focus, leaving concepts such as engagement or culture with an infinitesimally small fraction of resources. [1]
The effectiveness of mass-scale investments in the 'I' domain is questioned by academic research. Studies cited in the analysis indicate alarming failure rates: half of senior employees resign or are fired within the first 18 months (Bauer and SHRM, 2010), and 50% of hourly workers leave the company within the first 120 days. Additionally, a study by Saks and Belcourt (2006) reveals that two-thirds of those who underwent training no longer apply the skills learned one year later. [1]
Performance management, a central pillar of HR tech, presents major deficiencies. Research by Scullen, Mount, and Goff (2000) shows that between 53% and 62% of the variance in feedback reflects traits of the person providing the feedback, not of the person receiving it. Moreover, Kluger and DeNisi (1996) found that feedback can negatively affect performance in more than a third of cases. The irony lies in the fact that the measurability bias is precisely the mechanism that sabotages the effectiveness of technologies built on these incomplete metrics. [1]
The structural problem targets what is not measured versus what is measured: selection evaluates the volume of applications and time-to-fill, not long-term results; training measures completion and satisfaction, not knowledge retention; performance management is often distorted by managers' desire not to demotivate staff, deviating from its true purpose of improvement. Results of hiring and training are difficult to monitor due to time costs, leading to an illusion of control based on input rather than outcome. [1]
However, the study identifies a possibility for recovery: artificial intelligence (AI) can be the tool that transforms qualitative data into demonstrable evidence of impact and ROI. Unlike the traditional HR approach that avoids the unquantifiable, AI can analyze large sets of unstructured data (such as anonymous feedback or team discussions) to extract measures from organizational psychology. Examples include using AI to summarize qualitative data in real time, creating reliable sources of organizational psychology insights, and facilitating team decisions by identifying conflicting viewpoints before discussions. [1]
Practical applications of AI in HR go beyond automating administrative tasks, aiming to reconnect with the human nature of work. By combining document analysis with assisted interviews, organizations can capture the real manner in which work is executed. Although the impact of AI on the future of work may be perceived as frightening, its potential to provide clarity in qualitative domains could lead to a future of work that is more human-centered rather than focused on bare metrics. Market data mentioned earlier (451 Research, HRO Today) is reiterated to highlight the widening gap between investments in 'I' and 'O'. Less than 2% of the $94 billion market for HR solutions is allocated to qualitative constructs, while $3.75 billion in 2025 investments went exclusively to the quantitative segment. The 451 Research analysis established the benchmark for investment divergence, highlighting that the HR tech industry focuses disproportionately on streamlining standardized data flows, ignoring human behavioral complexity. Observations regarding the strict focus of human and financial resources on the 'I' side of the I/O spectrum are reinforced by investor tracking reports. The 'O' sector, including culture and leadership, remains a critically uncovered area of current technology, leaving organizations vulnerable to decisions based on incomplete data. The central thesis of the article, concerning uncertainty as the driver of the measurability bias, is reaffirmed through the lens of organizational behavior. Academic research regarding the invalidity of feedback (Kluger, DeNisi) and hiring failure rates (Bauer, SHRM) are reintroduced to validate the premise that 'I' does not function correctly. Longitudinal studies cited (Saks, Belcourt) show the inability of traditional training to maintain long-term behavioral change. The ability of AI to process unstructured data represents the necessary break from measuring procedures to measuring human outcomes. Modern sentiment analysis and summarization tools can transform HR updates from 'compliance tracking' to 'understanding impact.' The future of HR tech appears to be defined by the hybridization of old metrics with new capabilities of linguistic algorithm-assisted qualitative analysis. [1]
- Categories:
- Economy
