The bait, then the rug-pull.
The pitch is the dream every solo operator has been sold a hundred times: more revenue, same hours. The proof is more interesting — one of Rashid's clients, a fractional CFO named Kieran, swapped 40 hours of monthly reporting for a 10-minute review pass and doubled his client list without hiring. The mechanic, and the slides Rashid uses to teach it, are what's worth studying.
What the video promised.
stated at 01:05“In this video, I wanna walk through a case study of my client, Kieran, a fractional CFO, who was spending forty hours a month on monthly reporting for his clients and got that down all the way down to ten minutes using a skill system.”delivered at 08:34
Where the time goes.

01 · The promise
Service owners are capped by calendar hours; AI as a skill system can lift the cap without more hours or hires.

02 · Revenue up, hours flat — the dream
Visualizes the 'two-curve' goal: revenue climbs while hours hold or fall. Sets up the diagram that the next chapter inverts.

03 · The trap — hours climb with revenue
Most owners using AI as a tool watch both curves rise together; they just work faster, not less.

04 · AI as employee, not tool
The perspective shift: treat AI like a hire you train over time, share IP and context with it, expect skill compounding.

05 · Skill systems defined
A skill system is a chain of skills AI runs as a workflow; improving any one skill is like coaching an employee.

06 · Meet Kieran
Fractional CFO for SaaS Series A/B founders, 7 clients capped at 6–8 hrs/client/month, $24K–30K/mo, turning away leads.

07 · Same ceiling you're in
Before/after card: 7 clients → 15 clients, 6–8 hrs → 10 min, $24K → $52K, hire zero. The result is shown before the method.

08 · This is what infrastructure looks like
The 8-step pipeline diagram: /run-monthly [client] → Load Config → Pull Data → Calc Metrics → Validate → Gen Charts → Gen Commentary → Render HTML → QA → human review.

09 · He knew. He just couldn't fix it.
Identifying the bottleneck is easy; carving the time to systematize it while still delivering is the actual blocker.

10 · Map first. Build second.
Used a business map + bowtie funnel to confirm the bottleneck was delivery, not lead gen — that determines build order.

11 · Build for one. Template the rest.
Get one client's pipeline to 90% quality (took ~40 hrs), then 70% of the work transfers to every new client.

12 · Recurring vs one-time cost
8–10 hrs/client/month forever flips to one ~10 hr onboarding, then zero hrs/month after.

13 · AI does the work around your judgment
AI builds the deck and flags anomalies; Kieran writes the commentary and decides what to say. Judgment stays human.

14 · Three paths
Stay capped, hire an analyst, or build a skill system. Only path 3 raises the ceiling — at $200/mo on Claude Code.

15 · CTA — Chief Leverage Officer community
Hard pitch: community launching May 18 for owners who want digital asset systems, skill systems, AI employees. Link in description.

16 · Recap calls → Strategy calls
Second-order win: Kieran sends a Loom of the numbers before the call, so the live call becomes strategy. Clients value it more.
Visual structure at a glance.
Named ideas worth stealing.
AI as tool vs AI as employee
Treating AI like a tool means you work faster and hit the same hour ceiling. Treating AI like an employee means you train it on your IP over time and let it compound — hours drop while revenue rises.
Skill System (chained skills)
- Load Config
- Pull Data
- Calc Metrics
- Validate
- Gen Charts
- Gen Commentary
- Render HTML
- QA Review
A skill system is a series of single-purpose AI skills chained into a workflow, triggered by one slash command, ending in a human review pass. Each link can be improved independently like coaching an employee.
Build for one, template the rest
Pick one client/use-case, build the full pipeline end-to-end to 90% quality, then 70% of the work copies to every new client with 20–30% customization.
Bowtie funnel for finding AI opportunities
- Traffic
- Converters
- Products
- Funnels
- Math
- Team
- Goals
Map the whole business (lead-gen left side, delivery right side) on one canvas before deciding where to apply AI. Most owners assume the bottleneck is leads; for service businesses it's usually delivery.
AI does the work around your judgment
Let AI handle the 80–90% repetitive execution (build the deck, flag anomalies, draft commentary). The owner does the 10–20% that requires judgment (what the numbers mean, what to tell the client). Don't outsource judgment.
Three paths to scale a service business
- Do nothing (stay capped)
- Hire an analyst (more management overhead, same ceiling)
- Build a skill system / AI employee (ceiling lifts)
The classic three-choice framing — only one option breaks the ceiling. Each option gets its own color-coded card (red/red/green).
Lines you could clip.
“Kieran was spending forty hours a month on monthly reporting for his clients and got that down all the way down to ten minutes using a skill system.”
“If you use AI like a tool, you're just gonna work faster. Yes, you might be able to serve more clients, but because you're working faster, you're gonna work more hours.”
“The perspective shift we wanna make here is we want AI to work as an employee.”
“Within two weeks of implementing the skill system, he was able to serve 15 clients.”
“Now he can make around 52k per month, and he doesn't need to hire anyone.”
“AI should do most of the work, but you should do most of the judgment.”
“It works on a $200 a month AI subscription on Claude Code.”
How they spent the runtime.
Things they pointed at.
How they asked for the click.
“I am launching a community next week on May 18. It's called the Chief Leverage Officer community... if you want that, I'm gonna put more information in the description below.”
Soft, value-led — lands after a full lesson, framed as the next step for owners who already buy the 'skill system' thesis. Re-mentioned at the close with no price reveal, deliberately driving to the description link.
Word for word.
Steal the slide rig, not just the thesis.
Rashid's slides are doing 80% of the teaching — the talking head is just a presence cue. That ratio is the unlock for technical content.
- Build 5–7 hand-drawn lesson cards (Excalidraw works fine) as the main canvas — drop yourself into a PiP bubble in the corner.
- Number every lesson top-left (`## · LESSON #`) so the viewer always knows where they are. Add a thin rail along the bottom that names each visual element on the slide.
- Color code religiously: green = the path you want them to take, red = the path they're on, white = neutral structure. Hand-circle the key number on every before/after table.
- Show the result before the method. Rashid drops the 7→15 clients / $24K→$52K table four minutes in, then spends the next twelve explaining how. Curiosity stays high the whole way.
- Every claim gets a 2D chart. 'Revenue up, hours flat' is literally drawn as two lines. Visual proof beats verbal proof.
- Pitch your offer (or product) as path 3 of 3 — make 'do nothing' and 'do the obvious thing' explicit, then show your option as the only one that breaks the ceiling.
- For Mod Boss / JoeFlow content: take any creator workflow (record → cut → caption → schedule → post) and draw it as an 8-step labeled pipeline. That single slide is the whole pitch.
If you run a service business, here's what to actually try.
Before you hire your next person, audit the one deliverable that eats the most hours every month — that's where AI can do real work for you.
- Pick the single recurring deliverable that takes the most repeated hours (monthly report, audit, recap, onboarding doc).
- Map it as a sequence of small steps: load inputs → pull data → calculate → validate → draft → render → review. Each step should fit in one sentence.
- Build the full pipeline for ONE client first. Don't try to generalize. Aim for 90% quality, not 100%.
- Use Claude Code or a comparable agent so you can keep the steps as files you own and edit — not buried inside a SaaS prompt box.
- Treat the system like a new hire: when it makes a mistake, edit the relevant step instead of just fixing the output. That's how it compounds.
- Keep the judgment work. Let AI build the deck and flag anomalies; you decide what the numbers mean and what to tell the client.
- Onboarding the system to a new client should take a one-time block (Kieran's was 10 hrs). After that, the monthly hours go to near-zero.
































































