Skip to main content

The Tight Five

· 7 min read
Dreamineering
Engineer the Dream, Dream the Engineering

OS Module: Architecture — The schema that organises schemas

The anchor of The Tight Five series


How many things can you hold in your mind and still act?

Not think about. Not write down. Not save to a spreadsheet that you'll never open again. Hold in working memory while making a decision under pressure.

The answer, for nearly everyone, is five.


The Felt Problem

You've been in the meeting. Twelve strategic priorities on the whiteboard. Everyone nods. Nothing changes. You've written the plan — forty-seven action items, colour-coded, time-boxed. By Thursday it's dead. You've had the insight at 2am that connects everything, and by morning it's fog.

This isn't a discipline problem. It's an architecture problem.

Your mind has a loading dock, and it fits five crates. Try to load six and one falls off the back. Try to load twelve and you're standing in a car park full of scattered cargo wondering where to start.

Miller's Number says seven, plus or minus two. But Miller was measuring recall, not agency. Recall is passive — you can hold seven digits long enough to dial a phone. Agency is active — you need to hold the elements, see their relationships, and decide what to do next. That costs more. Five is the budget.

The question is: which five?


Why "Tight"

The word carries three loads. All three matter.

Bound. In rugby, the tight five are the two props, the hooker, and the two locks. They bind together in the scrum — physically, literally, arms interlocked under eight bodies pushing against eight bodies. Remove one prop and the scrum collapses. Not weakens. Collapses. The tight five don't score tries. They create the platform that makes tries possible. Every flashy backline move in the history of the game started with five people doing unglamorous work that nobody films.

Polished. In stand-up comedy, a "tight five" is a comedian's best five minutes. Not their shortest five minutes — their most refined. Comedians perform hundreds of times, in rooms that don't laugh, cutting every word that doesn't land, adjusting every pause, until what remains is pure signal. A tight five sounds effortless because everything that required effort has been removed. That's why the best pitches feel like conversations — they've been compressed until nothing remains that doesn't serve the person listening.

Incompressible. Both meanings converge here. You cannot remove a prop from the scrum. You cannot cut a joke from the tight five. What remains is what must remain. The test isn't "is this good?" The test is "does the system collapse without it?"

The count isn't the point. The binding is.


The Swap

Here's the move most people miss.

The Tight Five isn't a fixed list. It's a slot structure. The five changes depending on what mode you're in, what domain you're operating in, what question you're trying to answer. The architecture is constant. The contents are contextual.

A Dreamer loads one five. An Engineer loads another. A Coach loads a third. Same mind, different schema, different world visible.

ContextThe FiveWhat You See
Building a businessPrinciples, Performance, Platform, Protocols, PlayersValue creation system
Checking your healthWairua, Hinengaro, Tinana, Whanau, WhenuaWhich wall is cracking
Preparing a pitchEthos, Logos, Pathos, Kairos, ToposWhere persuasion leaks
Examining your lifeWhy? What's true? What do I control? Who's with me? Is it working?Where you're lying to yourself
Switching modesDreamer, Engineer, Realist, Coach, PhilosopherWhich processing mode fits now

Scott Adams wrote that the ability to change your mind is one of the best life skills you can hope to develop. He's right, but incomplete. The skill isn't changing your mind. It's changing your schema — loading the right five for the context you're in, then acting from that frame instead of whatever frame you happened to wake up with.

The meta defines the matter. Whoever loads the schema controls what becomes visible.


The Commissioning Sequence

Order matters.

In process engineering, commissioning means verifying that a built system actually works. You don't test the control loop before the pipe is installed. You don't calibrate instruments before the wiring is proven. Each component progresses through states: idea → spec → drawing → procurement → installed → wired → controls proven → operating. Readiness is the weighted blend of all components at their different stages.

The Business Five works the same way:

PositionElementWhy Here
1PrinciplesWhat's true? Without this, every downstream decision is a guess dressed as strategy.
2PerformanceWhat does good look like? What does bad look like? Without this, you're building blind. Define good cheese before designing the pipework.
3PlatformWhat creates leverage? Only meaningful once you know what you're measuring and why.
4ProtocolsHow do things coordinate? Only valuable when built around performance targets, not vibes.
5PlayersWho does what? Only assessable against the system they operate in.

Most people start at Platform — they buy the tools before defining the problem. Or they start at Players — they hire before knowing what the job actually requires. Commissioning teaches you to resist that impulse. Wire the principles first. Prove the measurement second. Build the platform third. Document the protocols fourth. Then — and only then — assess who plays what role.

It's unglamorous. It's the tight five. It's what makes the flashy stuff possible.


When Mantras Graduate

Here's a connection that took me too long to see.

A mantra is a trigger phrase — the human equivalent of a hook. "Act on principles, not emotion." "Ship, then iterate." "One task, full attention." Good mantras. Useful under normal conditions.

But mantras fail under load. The mantra page on this site says it plainly:

LevelMechanismFails When
MantraYou remember the phraseCognitive load, fatigue, emotion
RuleContext loaded, you apply itPressure, context overflow
HookFires automatically on eventSystem misconfigured
SystemEnvironment prevents the errorNever (if designed right)

The Tight Five is what mantras become when they graduate to systems.

Instead of remembering a phrase, you load a schema. Instead of relying on willpower, you have architecture. Instead of "I should remember to check all five dimensions," you have a frame that makes forgetting impossible — because if any element is missing, you feel the gap the way you'd feel a missing prop in a scrum. The thing falls over.

FromTo
Remembering a phraseLoading a schema
Personal disciplineShared standard that compounds
Willpower under pressureProtocol that fires automatically
"I should remember to..."Platform that makes forgetting impossible

This is the upgrade path. Mantras → Rules → Hooks → Systems. The Tight Five is the system layer.


The Instrument

This isn't just philosophy. It's a product.

The Prompt Deck is a 5x5 alignment instrument. It helps people find their own five — not by telling them what matters, but by running them through the structure until their five emerges. Discovery level: Capture, Inquire, Design, Platform, People. Strategic level: Principles, Performance, Platform, Protocols, Players. Tactical level: 5 Facts → 5 Questions → 5 Answers → 5 Ideas → 1 Decision.

Two modes. Pep Talk — the inner loop, selling yourself first. Pitch — the outer loop, aligning others to your vision.

The first and most important sale is always to yourself. If your own five doesn't hold, no pitch will save you.

The Time + Mind instrument is the complement. Prompt Deck answers what matters. Time + Mind measures did you actually spend your attention on what matters. Together they close the gap between intention and reality.


The Fractal

The same five runs at four scales simultaneously:

ScaleQuestionInstrument
AgentAm I pattern-matching or thinking fresh?Prompt Deck
IndividualIs my reality aligned with my ideal?Time + Mind
TeamAre we seeing the same pictures?Flow Maps
OrganizationDoes each team have clarity of purpose?Scoreboard

Misalignment at any scale produces dis-ease. Harmony produces flow.

What works at one scale works at all scales — the five steps compress the same way whether you're an agent processing a task, a person planning a week, a team shipping a feature, or an organization setting quarterly priorities. Same architecture. Different contents.


The Loop

Static fives decay. That's the failure mode of every strategic framework ever built — Step 5 never happens. You create the artifact, it captures a moment of clarity, and then it freezes. The world moves. The artifact doesn't.

The VVFL is the antidote. Validated. Virtuous. Feedback. Loop.

Every Tight Five instance must revolve and evolve. The questions sharpen. The principles get tested. The protocols get stress-tested by reality. The platform absorbs what worked. The players grow. Each cycle compounds — not just maintaining the five, but upgrading it.

The master loop:

1. QUESTION   → What's the gap between intention and reality?
2. PRINCIPLE → What truths filter noise from signal?
3. PROTOCOL → What method closes the gap?
4. STANDARD → What threshold defines success?
5. PLATFORM → What infrastructure makes the next cycle faster?

BETTER QUESTION → Loop compounds

Standards compound. But only if the loop turns.


The Question

What five would you load right now?

Not abstractly. Not as a thought experiment. For the thing you're actually working on — the decision you're circling, the project that's stalled, the conversation you're avoiding.

Which five elements, if you held them in mind simultaneously, would give you the clarity to act?

And which one are you leaving out — the one that would change everything, the prop that would stop the scrum from collapsing — because it's uncomfortable to include?

Load it. Bind it. Act.

That's the tight five.


Context

Questions

What makes five the right number — cognitive science, or the binding that prevents removal?

  • If the architecture is constant but the contents are contextual, how do you know when you've loaded the wrong five?
  • Where does the commissioning sequence break — is there a domain where Performance before Platform is wrong?
  • The mantra-to-system graduation path assumes systems can be designed right — what happens when the system itself needs to evolve?