The impact of turnover on operating margin

Analysis of the hidden costs of training and new technological levers to reduce Time-to-Competence in the factory

Italian industrial production closed the year confirming a trend of structural contraction, with a drop in volumes that reached 2.1% in the first nine months of 2025 according to industry surveys. In a scenario characterized by stagnating demand, the defense of EBITDA no longer passes through the expansion of production capacity, but rather through the efficient saturation of existing resources. A paradox is however eroding the competitiveness of manufacturing: while orders slow down, the difficulty in finding and making technical personnel autonomous is hitting historic highs, transforming turnover into an unsustainable cost item for the corporate balance sheet.

The current operating context

The bottleneck is not technological, but demographic and competency-based. Deloitte's 2025 Manufacturing Industry Outlook highlights how the sector is facing a double pressure: the mass exit of the expert workforce and the simultaneous entry of new resources with an increasingly wide technical skills gap. It is not simply a matter of vacant positions, but of roles filled by personnel who do not yet possess full mastery of the processes. Aggregate data show that the average time required for a new operator to reach full autonomy has increased from three months to over six months for complex tasks in the last five years. This extension of training times weighs heavily on the income statement. In a context where unit labor costs increase due to wage inflation, having low-efficiency operators for a semester means sustaining fixed overhead costs against reduced marginal productivity.

Rockwell Automation's 9th State of Smart Manufacturing Report also confirms this criticality, indicating the lack of skilled labor as the main obstacle to growth for 83% of global manufacturers. The problem transcends Human Resources management and becomes an operational criticality that directly impacts OEE and the ability to meet delivery deadlines. Traditional training, based on job shadowing or reading paper manuals, proves to be an unsustainable model because it subtracts productive time from expert personnel engaged in mentoring and transfers skills too slowly.

Analysis of the economic impact

To understand the real extent of the problem, it is necessary to abandon qualitative training metrics and adopt the quantitative ones of management control. The real cost of a new operator is not limited to the company cost during the trial period, but is the sum of often overlooked financial vectors.

1. Opportunity Cost of Supervision: when a shift leader or a senior operator dedicates a significant portion of their time to training a new hire, the production capacity of the plant's highest value-added resource is reduced.
2. Costs of Poor Quality: during the learning curve, the frequency of errors, scrap, and rework is statistically higher than the plant average. Such costs should not be charged to the training budget, but are direct losses of margin that inflate the actual bill of materials.
3. Proliferation of non-standardised procedures, the so-called Shadow Processes: in the absence of immediate guidance, the inexperienced operator tends to develop personal methods to complete the task.

These deviations from the standard introduce a variance into the process that, in the long run, compromises the repeatability and reliability of the line. The National Association of Manufacturers in its Outlook emphasises how workforce efficiency has become the only lever for cost control in a market where input prices remain volatile. If a plant records an annual turnover of 10% and takes six months to make a replacement autonomous, the plant structurally operates below theoretical capacity for 5% of the year. In terms of the income statement, this translates into an inefficient absorption of fixed costs that penalises the operating result. Technology must intervene to compress the time that elapses between onboarding and full operational status.

Cost-benefit analysis and Organizational Prerequisites

The emerging technological response, identified by Gartner in its strategic trends for 2025, is the adoption of Agentic Artificial Intelligence. Unlike traditional generative systems that provide passive textual information, AI agents are designed to understand the operational context and guide the operator step by step.

From research to resolution

The integration of AI agents into operators' devices allows for the transformation of skills development from a one-time event to a continuous "on-the-job" process. The AI agent delivers the operating instruction at the exact moment it is needed, according to the principle of Just-in-Time Learning. Preliminary data on pilot implementations show a reduction in Time-to-Competence of up to 40%. Financially, this means recovering months of full productivity and reducing the opportunity cost of seniors taken off the line.

TCO and ROI Analysis

The adoption of these solutions involves a shift from CAPEX to OPEX. The Return on Investment is high in specific scenarios such as "high-mix and low-volume" productions, where format changes are frequent, or in high-turnover contexts where the onboarding cost is recurring. The investment loses attractiveness if the company operates in an extremely stable production regime with near-zero turnover and processes unchanged for decades, a situation in which technology costs might exceed the marginal benefits of efficiency.

Capturing tacit knowledge

A frequent obstacle to the adoption of advanced technologies is the poor quality of existing documentation. Many plants struggle with obsolete, fragmented or non-existent SOPs. However, the modern approach does not necessarily require rewriting tons of paper documents before starting. The real opportunity lies in the ability of AI to capture and structure tacit knowledge. Instead of waiting months for the formal drafting of procedures, the new platforms allow for the recording of video or audio of expert operators while they perform the task correctly.

The AI agent processes these raw inputs, automatically transforming them into step-by-step standard procedures, checklists and interactive guides for new hires. This reversal of the process allows for bypassing "documentary debt." The lack of written procedures ceases to be a block to entry and becomes the trigger to digitalize the company's tribal know-how, making it a transferable corporate asset no longer tied to a single person.

Strategic Conclusions

The analysis has highlighted how the hidden costs of turnover and the lengthening of training times are silently eroding industrial margins. The demographic gap cannot be filled with traditional methods, making efficiency in onboarding an absolute financial priority. In this scenario, the adoption of Agentic AI is confirmed as the only lever capable of radically transforming the learning curve of new hires.

However, the challenge for Plant Managers does not lie in mere technological adoption, but in the ability to capture and codify the wealth of tacit knowledge currently at risk. Overcoming this obstacle requires a paradigm shift: to stop training for memory and to start designing for real-time assistance.

In definitive, future competitiveness will belong to those who can make corporate know-how a liquid and immediately transferable asset. Because in a volatile market, speed of competence is the new hard currency.

Sources

• Deloitte (2024): 2025 Manufacturing Industry Outlook.
• Gartner (2025): Top Strategic Technology Trends for Manufacturing 2025.
• Rockwell Automation (2024): 9th State of Smart Manufacturing Report.
• Federmeccanica (2025): Analisi Congiunturale Metalmeccanica.
• National Association of Manufacturers (2025): Manufacturers’ Outlook Survey Fourth Quarter 2024.