When an experienced architect hands in their notice, the practice rarely worries about the drawings. Those are filed, backed up, sitting safely on the server. What quietly leaves with them is harder to see: the reasoning. Why the cladding spec changed after the second site visit. What the planning officer would and wouldn't accept. Which decisions were the client's, and which the practice talked them out of. That knowledge isn't in a file anywhere — it's in someone's head, and on their last Friday it walks out of the door.
The knowledge that walks out the door
Every project is two things at once. There's the record everyone can see — the model, the drawings, the issued documents — and there's the layer underneath it: the decisions, the trade-offs, and the reasons. A practice runs on that second layer. It's how a project manager knows which battles were already fought, how a senior knows why an odd detail exists, how anyone knows what the client actually meant.
The trouble is that this layer is almost never written down. It accumulates in the people who were in the room. When they leave, it goes with them — and the loss isn't obvious straight away. It shows up months later, when someone reopens the project, finds a decision that makes no sense, and has no way to ask the one person who knew why.
Why handovers don't catch it
A good handover helps, but it can only carry so much. In a week or two, someone tries to compress years of context into a document and a few meetings. They write down what they remember to write down — and they forget the rest, not through carelessness but because so much of what they know has become instinct. Ask an experienced architect why they made a particular call and the honest answer is often "I just knew." That instinct is exactly the knowledge the practice most wants to keep, and it's the hardest to hand over.
Handovers are also a snapshot. They capture what one person can recall on one day, near the end, when they're already half out the door. Everything they decided two years ago, on a project that has since moved on, is long gone.
It's scattered across meetings, emails and memory
Part of the problem is that the record of why is spread across the worst possible places to find it later. The real decisions happen in meetings and on site — spoken aloud, rarely minuted in any useful detail. The follow-up lives in email threads and chat messages that only a few people can see. And the connective tissue — the part that explains how one decision led to the next — exists only in the memory of whoever was paying attention.
So even when a practice technically "has" the information, getting it back means hunting through inboxes, opening half-remembered files, and asking around to see who might recall. When the person who recalls has left, that route closes.
What a record that survives turnover looks like
The practices that hold onto their knowledge tend to do one thing differently: they capture the reasoning as it happens, not at the end. Decisions are recorded close to the moment they're made, in plain language, alongside the context that explains them — who was there, what the alternatives were, what the client wanted. It doesn't have to be heavy. It has to be consistent, and it has to be searchable, so that two years later someone can ask a question and get an answer rather than a shrug.
Done well, this changes what leaving means. The drawings were never the problem. When the reasoning is part of the record too, a departure stops being a loss of knowledge and becomes what it should be: one person moving on, with everything they knew about the work still held by the practice.
A lighter way to keep institutional knowledge
This is the problem we're building lintil to solve. lintil captures what was decided in your meetings and site visits, why it was decided, and where it happened — and turns it into a record the whole practice can search. Nothing becomes part of that record without a person confirming it should; the aim isn't to replace anyone's judgement, but to make sure the practice keeps it.
We're working with a handful of UK practices to build it. If keeping knowledge through staff changes is something you've felt the cost of, we'd like you in early.