When to establish completion control: the FEED and detailed engineering window

Completion control has a timing paradox. Its consequences arrive in the final year of the project, so that is when it gets attention. But its foundations are laid in the first year, when nobody is thinking about handover: in how the system structure is defined, how the completion database is designed and populated and what the contracts require of every party that will one day claim work is complete.

By the time completion becomes urgent, those foundations are set. The project is then improving its completion outcome with the most expensive tools available: rework, task forces and negotiation. The projects that hand over cleanly made a different choice early. They treated the FEED and detailed engineering window as what it is: the one period when completion is cheap to get right.

What should be settled before contracts harden

Four things, none of them onerous, decide most of the completion outcome. The system and subsystem structure, defined and agreed with commissioning input, so the project is built around the units it will hand over. The completion database, designed and populated against that structure, with tag registers loaded and inspection and test requirements allocated before the volume arrives. The inspection and test strategy, aligned with the quality plans of every contractor, so evidence is generated once, in the right format, rather than reconstructed later. And the contracts themselves, which should oblige every party to a common completion language: the same boundaries, the same records, the same gates.

Settled early, these cost meetings. Settled late, they cost months.

The early assurance question

There is also a role for independence in this window. Once the completion framework is designed, populated and connected to the contracts, an independent check answers a question that is almost impossible to ask from inside: will this structure withstand execution? Are the boundaries commissionable, is the database coherent, will the evidence disciplines survive contact with a construction ramp-up? It is a short engagement at this stage, and it is the highest-leverage assurance money the project will ever spend, because everything discovered costs a revision instead of a recovery.

Completion is not a phase. It is an outcome the whole project either builds towards or discovers. The window for building is open now, and it closes quietly.