0 À 100K€/JOUR, CLAUDE AI, RICHESSE, 1.5M€/MOIS | MASTER #58

Business strategy & operating model (what scaled)

• Team + delegation via KPIs (“mini-entrepreneurs”)

  • Each function owns a department and is managed through clear KPIs, not day-to-day micromanagement.
  • Weekly operational health reporting is centralized through a Head of Operations, who collects metrics from:
    • Customer Experience (CX) → customer satisfaction
    • Inventory / inbound stock → status for 3PL in the US
    • Warehouse / ops → operational readiness (Europe uses rented space; US uses 3PL currently)

• Scale via LTV/CAC discipline for subscriptions

  • Decisions are driven by unit economics across rebills, not by single-platform “profitability.”
  • They allow ad platforms to be temporarily less profitable if they bring customers at acceptable M/CAC targets once retention is included.

Frameworks / playbooks explicitly referenced

• Pareto (80/20)

  • Scaling requires finding leverage and pushing only the 20% of actions that create most output (e.g., “best funnel + best creative”).

• LTV/CAC calculator (time-based payoff)

  • Track profitability at 30/60/90/120 days (and longer) to determine allowable acquisition spend after payback.

• Unit economics thresholds (rebill-aware)

  • A ROS-based target drives ad spend pacing (subscription economics).

• “Comparator” / competitive teardown process

  • Mirror competitor email flows and reverse-engineer funnels.
  • Also use placements via ordered carts on mobile to uncover real flow depth.

• AI-enabled internal creative/ops systems

  • Internal platform with modules such as:
    • analyzer
    • script writer
    • studio/static generator
    • end card library
    • landing page generator
    • learnings/feedback loop
    • leaderboard

Key KPIs, metrics, and targets mentioned

Growth / scale

• Internal growth markers: “30 to 50 cadets… from 50 to 100…” (“cadets” = creatives/leads/content output in their internal language)

• Team growth: solo → ~22 people

Ads & creative performance

• Creative mix shift:
  • ~60% video / 40% native (later still ~40% native)
• CTR treated as a primary North Star:
  • Higher CTR = more optimization opportunities and protects account performance.
• Applovin-specific
  • Emphasis on end cards (N-cards) due to outsized influence on CTR and performance.

Subscription unit economics

• ROS ranges (internal metric)
  • Spend pacing based on ROS around 0.70–0.75 and 1.1
  • When ROS is ~1.1, they’re willing to spend more
  • When ROS drops toward ~0.75, they slow spend
• Payback timing
  • They re-bill every ~2 weeks
  • Customers are charged 3 times in the purchase month
  • Result: faster payback than typical 30-day models
• LTV/CAC
  • They compute LTV/CAC and use it to set allowable acquisition spend over time (30/60/90/120 days).
• Subscription rate target
  • Subscription rate target around ~75% (used in offer economics logic)

Amazon / retail entry thresholds (execution rule-of-thumb)

• “Try to position yourself on Amazon once around 150k–200k/month”
• “In the US, around 200k–300k” (then expand to retail afterward)

Concrete examples & case studies

Skincare subscription cohort risk (“not all audiences are created equal”)

• Example audience/brand: Er Paris skin niche “GLP” (targeted audience undergoing drastic transformations)
• Observations:
  • ROS was strong early (example: spend ~20k on audience; ROS 3; ~60k on that CBO)
  • Later cohorts worsened: chargebacks and churn increased quickly
• Root-cause explanation:
  • Audience expected transformation didn’t match product results timing → disappointment after first trial
  • Another audience with aspirational + slower transformation retained better due to:
    • placebo/visual progress
    • lower disappointment

Partnership ads with a “jackpot” incentive mechanism

• Goal: enable creators to make aggressive claims.
• Mechanism:
  • Offer a commission/payout “pot” (example cited around $40 pending) unlocked by performance thresholds
  • Structured so creators can’t withdraw immediately
• Outcome:
  • Creators post more frequently and use more persuasive messaging (chasing the unlock)
• Framing:
  • “Like sports betting deposit/win-to-unlock”

Competitor teardown example: mobile upsell visibility

• To analyze upsells accurately:
  • Place an order on mobile
  • Screen-record to capture the full upsell funnel (desktop may hide depth)
• Example mentioned: Cupid Cupid perfume brand
  • Mobile upsell flow increased AOV by about 5x
  • Attributed to funnel mechanics + checkout fields

Go-to-market / growth tactics (ads, funnels, subscriptions)

• Creative strategy evolution

  • Due to saturation on Meta:
    • moved from simpler formats to longer VSLs
    • increased more organic execution
  • Uses street interviews as a proven ad format

• Ad platform strategy

  • Core spend:
    • Meta (still major)
    • Google (slightly less performance but still spend)
    • Applovin (works well; explored for scale)
  • Applovin creative approach
    • Create creative “in Applovin mode,” not copied 1:1 from Meta winners
    • Applovin audience tends to be older, with emphasis on women 40–50+
    • Mobile reward-based viewing behavior is important

• No cutting ads

  • Generally don’t block creatives (unless major conditions)
  • Incremental attribution remains profitable
  • Optimize using daily/overall subscription revenue (MR)

• Offer strategy for subscriptions

  • Target consumables aligned to daily routines (e.g., toilet paper, soap)
  • Caution: avoid product types with low “mandatory consumption” (e.g., some supplements)
• Upgrade/upsell timing
  • Use upsells optimized for longer periods to:
    • collect cash earlier
    • reduce payment friction
• Funnel content strategy
  • Shift in native writing length:
    • older: ~2500-word writing
    • new: ~4000-word writing
    • selling occurs between 4000–5000 words
  • Many audiences are now problem-aware / product-aware, not fully cold (due to saturation)

Operations & systems

• Warehousing / logistics

  • Europe: rented warehouse space for one brand
  • US: currently 3PL
  • Plan: build/obtain own warehouse by 2027 (savings + control)

• Weekly operational reporting

  • Central summary from Head of Ops using department-level data.

Payment processing & risk management

• Shopify Payments

  • Kept for simplicity once scaled.
  • Onboarding risk acknowledged early (“gamble” in Shopify risk assessment).

• Chargebacks

  • Improved by US stock + faster experience.
  • Warns about high chargebacks when fulfillment timing causes “rebill before customer can use product,” especially for sourcing/shipping delays.

Practical actionable recommendations (distilled)

• Build “mini-entrepreneur” teams

  • Give each role KPIs and reduce founder involvement to weekly dashboards.

• Optimize for subscription economics, not per-platform vanity

  • Use LTV/CAC and ROS targets, considering rebill timing.

• Treat end cards (N-cards) as CTR levers

  • Especially on Applovin: focus on card CTR improvements.

• Use mobile teardown for upsell accuracy

  • Order on mobile and screen-record the post-purchase funnel.

• Segment audiences by LTV quality, not only initial ROS

  • Run cohort-based analysis to avoid assuming volume equals retention.

• Use competitive email-flow mirroring

  • Scrape and compare email sequences; annotate timing (e.g., day after purchase, 1 week after).

• Implement an internal creative/ops knowledge base

  • Use AI to standardize workflows:
    • analyze ads → generate scripts/statics → create landing pages → store learnings.

Presenters / sources mentioned

• Loï (main guest; communications and multi-brand operator)
• Nico (host / co-presenter)
• Luc (appears near the end; mentioned in relation to school/community/podcast)
• Mathieu (mentioned during CPM discussion)
• Claude Code (tool referenced; implemented via Grock/AI workflows)
• Grock / Grockbot (AI tool referenced)
• Gemini Flash (AI tool referenced)
• Miro (used for competitive teardown workflows)

Competitor / brand examples referenced

• Fennel, Rosabella, I’m8, Er Paris, Resilia, Gruns, L’foria, Primal Queen, Imforia (spelled variably), Cupid Cupid

Additional people referenced

• Roman Kh, Lauris, Tom Nico (named in Applovin context)
• Philippe Lee, Sharan Sriatsa (acquisition.com reference)
• Hormosi partner context (Sharan mentioned)
• Martin (podcast inspiration reference)
• “Adam/Matthéo” (retail/Amazon/China context; name appears as Matthéo)