CLAUDE AI, 0€ À 1M€ PROCESS, RECHERCHE AVATAR, META ADS, SPY TOOLS | MASTER #55

Business-focused summary (strategy, operations, marketing/sales, execution)

1) Operating thesis: growth is amplified by “environment + internal programming”

The hosts argue that your business execution outcomes are strongly influenced by:

• Environment: being surrounded by ambitious, high-velocity, high-expectation peers (their example: Dubai vs. Switzerland/Europe).
• Internal programming: how you think, apply advice, process feedback, and manage emotions.

Actionable implication: Don’t only optimize tactics (ads/creatives)—ensure you’re in an environment that supports fast learning and ambition.

---

2) “Invest in yourself” as the highest ROI lever (capital allocation principle)

They recommend allocating a portion of savings (described as a “portion” that can be 1x / 10x in cases) to self-improvement, including:

• Training, coaching, mentoring
• Self-development
• E-commerce mentoring

Framing: ROI from skill-building compounds more than financial instruments (they explicitly contrast against ETFs/real estate).

---

3) Visualization + mindset = a measurable execution prerequisite (inputs affect outputs)

Before “technical” marketing, they push a process of changing how you represent money internally:

• If you visualize money incorrectly (e.g., “digital/coins that don’t feel real”), you may lose gains repeatedly and not extract emotional/behavioral value.

They describe this as neurolinguistic programming (NLP-style) and claim it changes behavior—and therefore results.

---

4) Core e-commerce marketing playbook: avatar → angles → concepts → creative testing

The largest technical section is a repeatable framework for creative strategy and ad production.

Framework / playbook

1. Identify the avatar(s) Person/segment with habits, constraints, pain points, and values.

2. For each avatar: Map angles Messaging perspectives that resonate with that avatar’s worldview.

3. For each angle: Build concepts + mechanisms

   • Mechanisms = why the product works for that avatar (not generic benefits).
   • Marketing blocks = psychological lever + timing/moment-of-life.

4. TAM logic One product can address multiple avatars, so you must focus on which avatar(s) and angle(s) actually convert.

5. Start narrow Use 1 primary avatar (or 2–3 avatars max) first, scale there, then expand angles after reaching strong baseline performance.

Concrete operational logic

• Their “FBI profiler” analogy: a creative strategist must know everything about the product + avatar to avoid irrelevant messaging.
• Suggested input sources:
  • Reviews/comments (e.g., Trustpilot) to extract:
    • emotional triggers,
    • language style,
    • high-intensity life moments (“when pain is intense”)
• Then translate triggers into ad structure (e.g., in weight-loss: the moment someone removes a shirt/pain point).

---

5) Competitive research system using spy tools + LLM workflow (Claude/“Cloud” + MCP)

They describe a technical stack that turns competitor intel into a ready-to-produce creative plan.

Workflow

• Use LLMs (they mention Claude and GPT in the cloud) plus spy tools.
• Connect tools via MCP (their “brain” layer that orchestrates tool actions).

Steps:

1. List competitors using meta/keyword-based discovery (They warn some spy tools may miss ~30%.)

2. Collect competitor creatives Analyze all creatives, including images/video hooks.

3. Generate deep competitor reports Identify what competitors:

   • print now,
   • use in angles/personas,

   • have abandoned. 4. Ask for prompt-driven output Focus on:

   • angles used 3–6 months ago (to avoid stale tactics),

   • what they don’t use today,
   • their new angle,
   • most-used angles in the last 30 days (especially with impression counts in Europe / filtering in the US).

Example prompt intent (described)

“Analyze this competitor in depth… identify:

• exact marketing angle + problem addressed,
• target avatar desires,
• segment awareness level,
• what angles worked previously and what stopped,
• current dominant angles over 30 days.”

Output

A large report that identifies:

• which creatives get the most printing volume,
• which competitors you should mirror vs. avoid,
• what ads your team should produce right now for printing/testing.

---

6) Spy-tool validation: how they infer scaling before metrics catch up

They address the problem where competitor shops may not appear in spy tool datasets yet are actively advertising.

Tactics / checks

• US “Shop Pay orders” hack (name not fully given, but described):
  • If the competitor is on Shopify and uses Shop Pay, you can see order count via a filter in the Shopify App Store on iPhone.
  • Use order volume to infer live scaling.
• SimilarWeb timing risk
  • SimilarWeb data can arrive around the 10th of the following month, so scaling may be understated.
• Cross-check higher-perspective multipliers
  • They claim SimilarWeb-reliant tools can later multiply visitor counts (e.g., 3–5x).
• Landing page (LP) coverage matters
  • Creative intelligence must check multiple LPs:
    • Missing pages can make active ad counts look too low.
  • Advice: ensure you view all competitor pages, not only the top page.

---

7) “Angle transfer” beyond direct product categories (indirect competitors)

They recommend expanding inspiration beyond direct substitutes.

Examples of indirect targeting:

• If you sell lip cosmetics, creatives that work for lips may translate to eyebrows (visual similarity).
• If you sell t-shirts, winning shirt angles can inform pants messaging.

This widens the angle search space so you’re not locked to “lip-to-lip” only.

---

8) Organizational operations: weekly blocker detection + automation ROI prioritization

Beyond marketing, they explain internal management and prioritization.

Process

• Every week:
  • employees note ad blockers or automations to pursue.
• Creative Strategy has a goal to reduce research time.
• They implement AI automation prioritization based on financial leverage.

Automation prioritization formula (conceptual)

Compare:

• automation cost (including paying AI experts at high monthly rates),
• time saved for lower-cost employees,
• payback window (they mention recoup in 7/10/15 months, depending on impact).

They emphasize:

• don’t build automations that are “cool” but don’t increase profit,
• prioritize tasks that generate leverage (time and/or direct profit).

Link to OKRs

Map tasks to whether they support company objectives (explicitly compared to OKRs):

• If a task doesn’t help earn money or avoid losses, it’s likely unnecessary.

---

9) Concrete internal tooling example: organizing creative production with a Drive-like system

They mention using a tool called R:

• “like Google Drive but smoother,” roughly ~$300 (approx.),
• to organize mess, speed editors, and improve output economics.

---

Metrics / KPIs mentioned (mostly operational/marketing-adjacent)

No formal company KPI dashboard is provided, but these quantitative signals appear:

Marketing intelligence KPIs / measures

• Active advertising signals from spy tools:
  • impression counts (Europe via meta library),
  • active ad counts,
  • printing volume / “most traffic.”
• Time windows for creative intelligence:
  • Angles used 3–6 months ago
  • Most-used angles last 30 days
• Competitor validation:
  • US order counts via Shop Pay
• Ad spend thresholds used as inference signals:
  • Ads struggling over 500k
  • Ads with 100k+ spend (used to infer creative structure patterns)

Internal operational ROI metrics

• Payback windows for automations: 7–15 months
• Employee cost comparison (examples):
  • €10,000/month expert vs. €2,000/month employee
• Scale numbers referenced indirectly:
  • building toward “over 100 million” (brand-building claim)
  • example revenue targets like “50 million/year” in the GPT role prompt scenario

---

Actionable recommendations extracted (directly from the hosts’ guidance)

• Pick the right environment early (speed of learning + ambition)
  • seek cities where you’re a “small fish among bigger wealth.”
• Allocate capital to skill acquisition first (train/coaching/mentoring).
• Use an avatar-first creative system
  • avatar → angles → concepts → mechanisms → creative blocks.
• Run competitor intelligence as a process, not ad-hoc research
  • connect spy tools via MCP,
  • validate scaling using multiple data sources,
  • check all LPs and not only top counts.
• Avoid stale angles
  • only keep angles that are still printing effectively; identify angles they stopped using.
• Expand via indirect competitors when direct copying stalls
  • borrow angles from visually/mechanically similar categories.
• Scale on 1–2 (max 3) avatars initially
  • get to strong baseline (implying at least ~1M/month with an avatar before expanding).
• Operationalize AI automation with ROI logic
  • weekly blocker collection,
  • prioritize automation by leverage and payback period,
  • map tasks to OKRs (earn/avoid loss), not just “innovation.”

---

Presenters / sources

• Nico and Matthéo (hosts/interviewees)
• Mentions/training/citations:
  • Robins Robins (referenced as teaching NLP)
  • Napoleon Hill (books suggested to feed into GPT)
  • Liquid Death / Alex (Liquid Death example; Alex referenced as head of gross)
• Tools/platforms mentioned:
  • Claude
  • ChatGPT / GPT
  • Spy tools including TrendTrack, BrainSearch, SimilarWeb, Meta Library, MCP
  • Shop Pay / Shopify App Store (US scaling validation method)
  • Trustpilot (comment source for emotional triggers)