The Handover Gap
The files landed on a Thursday afternoon. Drawings, schedules, equipment records. A model of the building as it was built. Someone from the project team walked the facilities director through the folder structure. The meeting lasted forty-five minutes.
Then everyone went back to work, and nobody was quite sure what to do with what they'd received.
I've been in that room. Not always on Thursday. But the meeting is always the same. Delivery is on schedule. The project is marked complete. And the model starts drifting from the building the moment the door closes.
The pharmaceutical facility was different. I was part of that one from the beginning, and the first thing we did was set aside the specification.
Not because it was wrong. Because nobody had explained what it was for.
We met twice a week with the facilities leadership team. Not to present. To understand. They were overwhelmed by what they'd been handed on previous projects — standards no one could maintain, data organized for the people who built the building rather than the people who ran it. So we asked a different question: what operational waste are you actually trying to eliminate?
The specification came out of those conversations. It looked different from what we'd started with.
When the project closed, they hired someone to maintain the database. Not because a contract required it. Because by that point, ownership was obvious. There was a person, a scope, a reason the data existed.
The model became part of how the building was managed.
I've also seen the other version.
A global organization. A well-designed, executive-backed specification for exactly the same problem. The model was accurate. The effort was genuine.
The site team hadn't shaped it. They knew it existed. They weren't sure what the data was actually for. When the project closed and friction appeared, they went back to spreadsheets. Not out of resistance. Out of gravity.
The enterprise thought deployment was complete.
The site kept working in Excel.
The model sat on a server.
The difference wasn't the technology. Both cases had accurate models and executive sponsorship.
The difference was who owned the record before the files landed.
In the first case, that question had been answered explicitly — by the people who would have to live with the answer. In the second, it had been implied. Assumed. Left to resolve itself at handover.
It didn't.
This isn't new. A checklist written for building owners in 2008 named data ownership and governance structure as explicit requirements — not recommendations. In 2015, a portfolio owner funded a person whose entire job was to maintain the accuracy of their building database. Not to operate the facilities. To own the record.
Both predate most of the platforms currently being sold as solutions.
The problem they were pointing at is still the same one.
The question worth asking before the files land: who in your organization is responsible for this data six months from now?
Not who will receive it. Who will own it.
If you can't name that person today, you're not delivering an asset. You're delivering something that will quietly expire.