Clarity WorksClarity Works

[ Core Memory ]

The operating library behind the method.

Every framework, concept, and mindset that keeps the YouTube channel, the Sprint, and client work pointed at the same thesis — published in full, because we charge for implementation, not information.

[ Capability Ladder ]
  1. 01AI fluency
  2. 02Business context
  3. 03Strategic thinking
  4. 04Workflow identification
  5. 05SOP improvement
  6. 06First internal AI-assisted build
  7. 07Capability transfer

[ 01 / Diagnosis ]

Understand the business and find where time and quality leak.

01original

Clarity Audit

A structured pass over the business to find AI opportunities worth pursuing.

Before any build, audit the operating reality: services, roles, recurring work, and where the owner is the bottleneck. The output is a shortlist of workflows scored for improvement — not a tool list.

02original

AI Business Context Document

The living source of truth that makes AI stop giving generic answers.

Services, customers, roles, workflows, constraints, and bottlenecks — written down once, reused in every AI conversation. The single highest-leverage document a team can create.

03original

Employee Context Document

The role-level version of the business context document.

Built around what a person actually does, how they work, and where AI can assist their specific tasks — so capability lands with individuals, not just 'the company.'

04original

Task vs Workflow vs SOP

The vocabulary that unlocks everything else.

A task is one unit of work. A workflow is the sequence. An SOP is the documented way it should run. Most AI confusion disappears once a team can tell these apart.

05original

Workflow Scoring Framework

Score workflows by frequency, risk, and payoff to pick the right first build.

The first AI workflow should be frequent enough to matter, low-risk enough to experiment on, and visible enough that the win builds momentum. Scoring keeps the choice honest.

[ 02 / Build order ]

The fixed sequence that keeps AI pointed at stable processes.

06adapted

5-Step Algorithm

Question, delete, simplify, improve the SOP, then AI-assist and automate.

Adapted from Musk's engineering algorithm, with one deliberate change: step 4 becomes 'improve the SOP.' Never optimize — and never automate — something that shouldn't exist.

07original

Capability Ladder

The seven rungs a team climbs from AI fluency to capability transfer.

AI fluency → business context → strategic thinking → workflow identification → SOP improvement → first internal AI-assisted build → capability transfer. Every program maps to rungs on this ladder.

08original

Optimization Order

Speed, then cost, then scalability — in that order.

Get the workflow fast enough to be used, then cheap enough to sustain, then scalable enough to grow. Reversing the order produces elegant systems nobody uses.

09adapted

10-80-10 Rule

Human starts, AI does the middle 80%, human finishes.

The first 10% (judgment, framing, context) and the last 10% (review, taste, accountability) stay human. AI earns the middle. Quality holds because ownership never leaves the person.

10original

Sprint Program

The 30-day activation that walks a team through the whole sequence once.

Context, diagnosis, SOP, build — compressed to one workflow in four weeks, so the team learns the method by shipping something real.

[ 03 / Offer & economics ]

How value, offers, and time economics get reasoned about.

11adapted

Value Equation

Value = (dream outcome × likelihood) ÷ (time × effort).

Hormozi's lens, applied to workflows: an AI assist is valuable when it raises the likelihood of the outcome while cutting the time and effort to get it. If it doesn't move one of the four, skip it.

12adapted

Grand Slam Offer

An offer so clear and de-risked that saying no feels irrational.

The Sprint's guarantees are this framework in practice: outcome named, timeline fixed, risk reversed on both the execution path and the service promise.

13adapted

Lead Magnet → Core Offer Ladder

Free value first, paid implementation second.

The content ladder behind this whole site: free YouTube frameworks → a free call → the paid Sprint → Partner. Every rung earns the next.

[ 04 / Adoption ]

How teams actually absorb AI capability without breaking.

14original

AI Capability = Fluency + Systems + Judgment

Capability is people who are fluent, systems that are built, judgment about where AI belongs.

Not engineers. Not tools. A team is AI-capable when its people can converse with AI usefully, a few well-built systems carry the repeatable work, and someone knows what not to automate.

15adapted

Buyback Principle

Buy back your time by cost, not comfort.

Martell's principle, pointed at AI: audit where the owner's hours go, and let AI plus documented process reclaim the lowest-value hours first.

16adapted

Pain Line

Owners delegate until it hurts — then take the work back and stall.

The Pain Line explains why AI adoption dies at the owner's desk. Crossing it requires process the team can trust, which is why SOPs come before automation.

17adapted

Replacement Ladder

Replace yourself in a fixed order: admin, delivery, marketing, sales, leadership.

Martell's delegation ladder, translated to AI: the same order tells you which workflows to AI-assist first — the ones lowest on the ladder, where risk is small and hours are real.

[ 05 / The rest of the memory ]

Mindsets, teaching skills, and the stories that carry them.

[ Mindsets ]
  • Problem-first, not tool-first
  • AI as a thinking partner
  • Sell transformation, not tools
  • First-principles thinking
  • Operating principles
[ Teaching skills ]
  • How to audit a business for AI opportunities
  • How to write a prompt that actually works
  • How to build a workflow with n8n
[ Stories ]

Phones vs. fax machines.

When a new operating layer becomes normal, the businesses that refuse it become harder to reach, slower to serve, and easier to leave behind.

The AI friend.

AI is like a smart friend who has read everything. The leverage comes from knowing what to ask and giving them the right context.

Both stories, in full →
[ Credibility rule ]

Proof has to stay honest.

The content engine separates real case studies from hypothetical examples. Public proof should name the mechanism, the workflow shape, and the measurable result without pretending a fabricated example was a client win.

Want this library applied to your workflows instead of read about?

Book a Call