The COO of a private logistics operator runs a network of twelve platforms across the Iberian peninsula: four large-surface distribution centers, five cross-dock platforms, three multimodal docks near ports. Closing the annual review with finance, the CFO points to an asymmetry that is hard to defend: operational surface area has grown twenty percent in three years; central operations headcount has grown forty percent over the same period. Both figures are correct. The first responds to the expansion plan the board approved. The second appears in no approved plan. It is the aggregate result of coordinating twelve platforms that operate as twelve semi-independent systems, and whose coordination consumes central-structure hours without producing measurable operations.
A logistics platform network is not twelve independent operations. It is a distributed system with shared costs, shared risks and shared learning curves. When it is managed as twelve isolated operations, coordination produces exactly the structural overhead that cannot be attributed to any specific asset or specific incident, and that nonetheless persists month after month.
Shared assets and the illusion of independence
Each logistics platform concentrates a fairly stable catalog of common assets: temperature-controlled warehouses, loading and unloading docks, forklifts and material handling equipment, HVAC systems, LED lighting, fire suppression systems, backup power, CCTV and access control, and increasingly rooftop PV structures and electric vehicle chargers. The typical size runs between one hundred fifty and four hundred common assets per platform; a twelve-node network therefore manages an aggregate portfolio on the order of three thousand elements. The operational illusion is that those three thousand elements are managed better as twelve blocks of two hundred fifty than as a single portfolio of three thousand, and the illusion rests on a single premise: that the platforms are different enough for cross-comparison to be uninformative. In practice, the differences are geographic and contractual, not technical: the same electric forklift behaves the same in Málaga as in Vitoria, and the same industrial refrigeration system degrades the same at both sites.
The three symptoms of dispersion
The first symptom is contractual duplication. Each platform contracts maintenance locally, without leveraging the network's aggregate volume. In a typical review, the same forklift model sits under three separate contracts with three separate service shops, at three different prices and SLAs, all twenty or thirty percent above what the network could negotiate centrally. The overspend appears in no single expense line: it is diluted across twelve entries.
The second symptom is preventive-plan drift. Each platform sets its preventive cadence from the local maintenance lead's memory, without cross-platform benchmarking. The result is preventive cycles that range from six to sixteen weeks for the same piece of equipment, with no technical justification supporting the difference. Neither the more frequent operation is measurably safer nor the less frequent one is measurably cheaper: they are historical artifacts nobody has renegotiated.
The third symptom is fragmented incident response. A repeated failure in a cold chamber at the Zaragoza platform is not communicated to Seville, which runs the same model and will hit the same failure within weeks. Each platform reinvents the lesson, and the cost of recurrence gets paid twelve times when it could be paid once. The cumulative effect of the three symptoms is not an abrupt overspend but a slow slope, one that becomes visible only when the board asks the operations lead to explain the growth of the central area, or when an acquired operation inherits the same asymmetry from the first quarter of consolidation.
The common base as a transverse layer
The operational answer is not to merge the platforms into a single center — that would be a strategic move outside the operations function — but to sustain beneath the twelve local operations a common layer: a georeferenced asset inventory with a single taxonomy, preventive plans defined as templates parametrized by platform, and a cross-platform comparison engine that surfaces MTBF, MTTR and cost per asset unit at the granularity central operations needs. Once that layer exists, the twelve platforms keep operating locally, but information flows upward with structure, and the central team stops spending time reconciling formats and starts spending it deciding which levers to move.
At Maptainer we work with this operator profile from an API-first architecture: each platform keeps its local SCADA, its WMS and its TMS, and the common asset management layer consumes telemetry from those systems via OpenAPI 3.0 and webhooks. Consolidation does not require migrating the tools of daily operation; it requires a clear data contract between what lives at the platform and what flows toward the central register. The central operator moves to seeing the network as a single distributed system, without breaking the operational autonomy of each node. The site manager keeps every lever they need for daily decisions, and the central function gains the vantage point that until now only appeared once a year, during the manual reconciliation exercise that consumed the first month of every fiscal year.
What the COO actually measures
The real KPI of the COO of a logistics network is not what happens inside a platform; that is the site manager's KPI. It is how the network behaves as a system: what the marginal cost of operating one more platform is, how long it takes a learned lesson to propagate between nodes, how large the gap is between the best- and worst-performing node on a comparable operational indicator. The moment the asset register unifies — one taxonomy, one geography, one comparison layer — the COO's job stops being to coordinate twelve operations and starts being to manage one distributed operation. That shift in framing is what makes the forty-percent growth in central headcount reversible without cutting service at any platform, and it is the conversation the COO can bring to the board with data rather than hypotheses. The next expansion cycle stops being an assumption of proportional headcount growth and becomes a defended plan under the assumption of decreasing marginal cost per platform, which is the shape in which acquisitions and organic growth start to compound the value of the operator rather than dilute the productivity of its central function.