The centralization of decision-making in companies is based on a simple principle: to escalate decisions to a small core of leaders or committees. This model, inherited from the industrial organizations of the 20th century, still structures a significant part of SMEs and large French companies. Several sector analyses point to an increasing gap between this decision-making rigidity and the rapid adaptation demands imposed by volatile markets.
Shadow systems: when data centralization has the opposite effect
Consolidating information into a single system (data lake, centralized ERP, cloud platform) aims to ensure data quality and simplify management. On paper, the logic is sound. However, field reports tell a different story.
See also : The puff: fashion trend or revolution in the vaping world?
Reports on the proliferation of shadow IT describe a recurring mechanism. When local teams do not find their needs met in the model imposed by the central platform, they recreate parallel databases and out-of-system files. Shared spreadsheets, personal folders, third-party tools not referenced by the IT department. The system intended to reduce fragmentation ends up amplifying it.
The consequences are direct: duplicates, inconsistencies between versions, risks of non-compliance with GDPR. For purchasing teams or inventory managers, operating with data whose reliability varies by source is like navigating without instruments. As analyzed by the centralization method according to Jeune et Actif, the gap between the process designed by management and the reality of daily usage generates more friction than the tool itself.
Related reading : Recent Changes in the World of Automotive Forums
Decision-making delays and loss of competitiveness in volatile markets
The health crisis served as a real-world test. Organizations that imposed systematic hierarchical validation for every adjustment in production, supply, or customer communication saw their reaction times increase significantly compared to structures that delegated decision-making autonomy to field teams.
This observation is not limited to multinationals. SMEs that centralize operational decisions around a single leader experience the same bottleneck. Each decision adds an additional layer of latency in the value chain.
Centralized management holds as long as the environment remains stable and predictable. It falters as soon as supply chain disruptions, demand fluctuations, or regulatory changes require rapid adjustments. The complete cycle (information escalation, arbitration, descent) consumes time that more agile competitors use to act.
Remote work and distributed teams: the centralized model facing talent retention
Gallup surveys (2023) on engagement and CIPD studies (2023) on hybrid work converge on one precise point. Highly centralized models (imposed hours from headquarters, standardized tools without room for adaptation, decisions locked by hierarchy) degrade engagement and retention of expert profiles and younger generations.
The issue goes beyond comfort. Employees with decision-making autonomy over their work organization show measurably higher levels of involvement. In contrast, those subjected to constant centralized control tend to disengage, reduce their initiative, and eventually leave the company.
For human resources departments, centralized management produces a hidden cost:
- Increased turnover in high-value positions, with recruitment and onboarding cycles directly impacting team productivity
- Loss of operational knowledge when experienced employees leave without their expertise being shared within the organization
- Difficulty attracting profiles accustomed to flexible work environments, which reduces the available recruitment pool
A warning signal for middle management
Frontline managers find themselves caught between rigid central directives and teams demanding autonomy. Their role is reduced to relaying decisions they did not help shape. This frustration fuels disengagement specific to middle management, which is identified as a performance lever in most organizational quality models.
Progressive decentralization: the paths tested by companies
Decentralizing does not mean eliminating all coordination. Companies that achieve documented results work on a clear sharing of decision-making scopes. Headquarters retains strategy, overall objectives, and data governance. Field teams gain autonomy over execution, the choice of tools suited to their context, and operational adjustments.
Several levers recur in recent feedback:
- Define explicit decision thresholds: below a certain amount or impact, the local team decides without escalation
- Implement federated data systems rather than a single data lake, to reduce the temptation of shadow systems
- Train middle management in autonomous decision-making, not just in applying top-down processes
- Measure the quality of local decisions by their outcomes, not by their compliance with the central process
The available data do not allow for a conclusion that a decentralized model works in all sectors or company sizes. In highly regulated sectors (healthcare, finance), centralization of certain processes remains a legal obligation, which limits the room for maneuver.
The main risk does not lie in centralization itself, but in its uniform application to contexts that do not need it. When a system designed for coherence produces slowness, parallel data, and disengagement, the first operational question is to determine which scope to decentralize as a priority.