A data center owner specifies 400 pieces of owner-furnished, contractor-installed equipment. The purchase orders go out. The manufacturers confirm ship dates. The construction schedule is built around those dates.

Six months later, a generator arrives at the jobsite with undocumented damage. Nobody knows when it happened. Nobody knows who touched it last. The vendor says it left the factory clean. The carrier has no record of an incident. The site crew is pointing fingers.

This is the accountability gap. And it costs the industry more than most owners realize before they've lived through it.

The OFCI model creates accountability risk by design

Owner-furnished, contractor-installed programs make financial sense. Owners can leverage their purchasing power, standardize equipment across a portfolio, maintain direct relationships with manufacturers, and build a data record that — if captured correctly — can improve how future programs are planned and executed. On paper, it looks like control.

In practice, most OFCI programs distribute that equipment across a chain of custody that nobody fully owns. A manufacturer ships to a freight forwarder. The freight forwarder hands off to a carrier. The carrier delivers to a warehouse. The warehouse holds it until the GC is ready. The GC coordinates final delivery. The installation crew takes it from there.

At each handoff, responsibility technically transfers. But accountability rarely does. Each party documents what they received, in whatever format their own system requires, and hands the equipment forward. If something is missing, damaged, or mislabeled, the discovery almost always comes late, usually at the point in the project when it causes the most disruption.

The data problem underneath the accountability problem

Most owners believe their OFCI equipment is being tracked. In many cases, it is; loosely, across multiple systems, with inconsistent standards and no shared source of truth.

A manufacturer has a ship date in their ERP. A carrier has a delivery confirmation in their TMS. A warehouse has a receiving record in their WMS. The owner's project team has a schedule in their construction management platform. None of these systems talk to each other. The data lives in silos, and nobody is reconciling it in real time.

When the project team needs to know where a piece of critical equipment is, they start making phone calls. When they need to know its condition, they hope someone took photos. When a discrepancy surfaces, they spend days reconstructing a chain of events from fragmented records.

That is not visibility. It is archaeology.

Research has found that US construction professionals spend an average of 14 hours per week on non-optimal activities such as searching for project data, correcting mistakes, and managing rework. That is 35% of their working hours. For the US construction sector at large, this accounted for over $177 billion in excess costs in 2018.

What the gap actually costs

The damage is rarely dramatic. It accumulates in the places that are hardest to measure.

A generator arrives at the jobsite before the pad is ready. The site team has nowhere to put it, so it sits on the trailer longer than planned. That costs money. When the pad is finally ready, the equipment needs to be repositioned. That costs more money. If it was sitting in an uncontrolled environment, there may be questions about condition. That costs time.

A switchgear unit ships with components missing — a configuration issue at the manufacturer that nobody caught on the way out. The project team discovers it during installation. Replacement parts have a 14-week lead time. The schedule absorbs the hit. None of these costs appear on a single line item. But they accumulate, and they all count toward the total cost of ownership for a program that looked predictable on paper.

A pallet of equipment arrives at a warehouse and is logged as received. When it is released to the jobsite three months later, three pieces are missing. Nobody knows if they were missing at receipt or left during storage. The warehouse record is a single line item. The investigation goes nowhere.

Each of these incidents has a specific cause. But the underlying condition is always the same: nobody owned that equipment at the moment the problem was created.

A joint study by PlanGrid and FMI found that roughly 48% of US construction rework is directly attributable to inaccurate, inaccessible, and incompatible project data, costing the industry more than $31.3 billion annually.

Chain of custody is a process, not a document

The common solution to accountability gaps is paperwork. Bills of lading. Receiving records. Inspection checklists. These create documentation, but documentation is not the same as accountability.

A receiving record captures what the warehouse accepted. It does not necessarily capture condition. A bill of lading records that a delivery was made. It does not capture what the driver saw when they opened the trailer. A checklist completed quickly and filed away creates a paper trail that is almost impossible to use as evidence when something goes wrong months later.

Research bears this out. Project teams with centralized, easy-to-locate documentation are twice as likely to feel confident proving which party was responsible when a dispute arises. Teams whose documentation lives in email threads, text messages, and phone calls report the lowest confidence. Documentation structure determines whether it can actually be used.

Real chain of custody means every touchpoint captures the same data — condition, serial numbers, quantities, photographs — in a format that feeds a single record the owner can actually use. Exceptions get flagged in real time, not discovered months later. And one party owns the continuity of that record across the entire chain, not just their piece of it.

Without that single point of accountability, the chain is only as strong as its weakest handoff.

What owners can do

The answer is not more auditing after the fact. It is building accountability into the supply chain before the first piece of equipment moves.

That starts with establishing a single source of truth: a system that captures and validates equipment data at every touchpoint, from manufacturer to jobsite, regardless of how many parties are involved in between. Serial numbers. Condition at receipt. Photographs. Inspection results. Schedule status. Unified data, accessible in real time, without having to request it from five different vendors, is what makes ownership of the supply chain real.

It also requires a standardized handoff process. If each carrier, warehouse, and delivery crew operates by their own documentation standard, the records will never reconcile. The data fields need to match. The exception flags need to be defined in advance. The process needs to be the same whether the equipment is moving through a warehouse in Virginia or a storage yard in Colorado.

And it requires that someone takes responsibility for the program as a whole, not just their leg of it. An owner who delegates logistics execution to five separate vendors has also, in practice, delegated accountability to nobody.

The question worth asking before the project starts

Before a single purchase order goes out, owners should be able to answer one question: if a piece of equipment is damaged between the factory floor and the job site, will I be able to determine exactly when and where it happened?

For most OFCI programs today, the honest answer is no. The records exist in pieces. The chain of custody has gaps. The accountability is distributed across parties with no obligation to share data in a format the owner can use.

The projects that handle this well are not the ones with the most rigorous documentation requirements after the fact. They are the ones that built accountability into the supply chain from the beginning: standardized data capture, continuous visibility, and a single partner responsible for maintaining the chain of custody from manufacturer to installation.

The equipment can arrive on time and in perfect condition, or it can arrive with a problem that nobody can explain. The difference is almost never luck.