Keep Five Divisions and Two Countries in Sync Without a Standup

The challenge

Code Four is not one business. Design, fabrication, production, project management, logistics and sales each run on their own rhythm, with their own rules and their own vocabulary, and a single event needs all of them to land on the same date.

Coordination fell to people. When fabrication finished a scenic piece, somebody had to tell logistics, who had to tell technical about the install window, who had to tell the project manager. Each handoff was an email chain or a standup, and each one was a place for a job to sit unnoticed for two days.

Airshow season compresses it. Permits, performers, sponsors, vendors and deliverables all move at once, while ordinary client work carries on beside it. Vendor POs and invoices need reconciling against what was delivered, which is the kind of task done thoroughly in January and approximately in June.

Then the time zones. Part of the operation runs in Australia, so work lands overnight and waits for an overlap window that barely exists. Whoever opened their laptop first spent the first hour triaging somebody else’s night.

Underneath it sat the Monday review of the whole operation: what is stale, what slipped, which handoffs are at risk, which vendor numbers disagree, which permits are close, which performers are unconfirmed. It took two to three hours to assemble by hand. It happened when there was time, and there was often not time.

What we built

We mapped how information moves between divisions, where handoffs stall and what everybody wished happened without being asked. Then we connected the systems already in use, email and project management and file systems, rather than replacing them.

Handoffs fire themselves. Fabrication completes a piece, the logistics timeline updates, technical gets notified about the install window, the project manager is told. No standup, no chain, because the system knows which divisions depend on which.

Each division keeps its own rules. Drafting a sponsor deliverable, disputing a vendor PO and submitting a city permit are not the same job and do not follow the same conventions. The system applies the right context to each because the business wrote down what matters where.

The overnight is triaged before anyone reads it. What arrived from Australia is sorted, actioned or escalated by morning, so the day starts on today’s work.

Vendors reconcile continuously. POs and invoices get checked against what was delivered as they arrive.

The Monday review assembles itself. Stale items, missed deadlines, at-risk handoffs, vendor discrepancies, permit deadlines and unconfirmed performers, in about thirty seconds. It runs every week now, because nothing depends on someone finding three hours.

The interface is whatever is nearest: text, email, Slack, voice memo, laptop.

What changed

Nobody is the middleware. Cross-division information moves because the system moves it, so the failure mode where a job waits on somebody remembering to forward an email is gone.

The weekly review became reliable. The check that catches problems while they are still cheap now happens whether or not the week allowed for it, which is the entire point of it.

The time zone became an advantage. Work progresses overnight instead of queuing for a two-hour overlap.

Airshow season stopped consuming everything else. Peak load is absorbed by the system rather than by pausing client work.

Why NimbleBrain

The instinct with an operation this complicated is to consolidate it onto one platform. That project takes a year, and event production does not have a quiet year.

We connected what the team already used and let each division keep working the way it works, then put agents on the specific jobs people were doing by hand: triage the overnight, reconcile the invoices, track the handoffs, assemble the status, draft the communications. Nothing was migrated and nothing was retrained.