Speedboat — Which Lane?

Use this when you’re not sure where a piece of work belongs.

Preview, Build, and Run are the work lanes. Land is the outcome layer.

Land is not a lane you plan into. It is a meaningful outcome: something reached a real beneficiary, or new evidence changed what happens next.

This guide answers what kind of work this is. It does not decide whether the work should enter the system immediately. Non-urgent new work should go into the unshaped backlog until Route Planning shapes it.

It also does not change based on whether a human or an AI agent did the work. Agent-assisted work still belongs in Preview, Build, or Run based on intent and risk.

START
  │
  ├─ Is this an urgent incident, bug, escalation, or emergency?
  │   └─ YES → **Run (Reactive)**
  │
  ├─ Is this planned maintenance, tech debt, dependency upgrade, or infra improvement?
  │   └─ YES → **Run (Proactive)**
  │
  ├─ Are we exploring whether we should build this? (Feasibility, appetite, unknowns)
  │   └─ YES → **Preview (Discovery)**
  │
  ├─ Are we making something demoable for Sales, a customer conversation, or stakeholder buy-in?
  │   └─ YES → **Preview (Demo)**
  │
  ├─ Is this committed, production-grade work with clear scope?
  │   └─ YES → **Build**
  │
  ├─ Has something just reached a real beneficiary, or did new evidence change what happens next?
  │   └─ YES → Log a **Landing**
  │
  └─ Still not sure?
      └─ It's probably a **Preview**. Start small, learn, decide Friday.

Quick Rules of Thumb

If you’re thinking:

“I don’t know if we should build this” -> Preview (Discovery)
“Sales needs something to show” -> Preview (Demo)
“We’ve decided to build this properly” -> Build
“This broke / this is urgent” -> Run (Reactive)
“We should fix this before it breaks” -> Run (Proactive)
“This reached a real beneficiary, or we learned something that changes what happens next” -> Landing

If you’re not sure, it’s probably a Preview.

Common Mistakes

Don’t put exploratory work in Build. Do treat it as a Preview first when there are still open questions.
Don’t let Run work stay invisible. Do make it a first-class card, tag it, and track it.
Don’t skip the Landing log. Do log it when work reaches a real beneficiary, or when new evidence changes what happens next.
Don’t let a Preview run for 2+ weeks without a decision. Do raise it on Friday and decide whether it is Build in disguise or should be killed.