CRÉAS IA, WHITELISTING, AD ACCOUNT, SCALE | MASTER #50

Business strategy & leadership takeaways

• High-leverage daily focus (“10 PORKs”): Put effort into activities that drive revenue (e.g., creative/testing/ads). Avoid getting scattered by every new tool.
• Mindset as an operating lever: Poor results are often framed as self-sabotage or a subconscious block. Discipline should feel “fluid,” not forced. Suggested interventions include coaching, hypnotherapy, and environment changes.
• Don’t micromanage; manage outputs: Hire experts, set clear objectives (e.g., more spend, better CPA, higher profit), provide processes, then supervise results—not every task.

Growth / marketing playbooks (Meta ads + creative strategy)

“Winning ad” definition framework (business KPI playbook)

Ads are evaluated by a combined set of conditions, not a single metric:

• KPI profitability
  • ROS must be above break-even (or CPA below target CPA).
• Spend threshold
  • A “true winner” should be able to spend meaningfully.
  • If it can’t spend beyond roughly $5K, it’s not fully validated.
• Time durability
  • The longer it lasts and continues spending, the stronger the signal.
• Scalability / account impact
  • Winners should drive account scaling, not just isolated performance.

Creative iteration & testing framework

• Winner iteration (voice / messaging variants)
  • If a winning creative targets men, iterate the same winner with more masculine voiceover.
  • For women, iterate with more feminine / sensual voice.
• “Premium 3D + black background” concept
  • Use 3D-style product visualization on a premium black background.
  • Pair with a “moon/microphone-like” launch-style voice for a luxury, scroll-stopping feel.
• Cartoon / Disney-style creatives
  • Continue using cartoon creatives (including in Europe where novelty is growing).
  • Double down because the effect may fade as novelty decreases.
• Rule of thumb
  • “Ads aren’t winning or losing—ads are more or less winning.”
  • Meta can show multiple ads at once, so performance is a spectrum.

Media buying scaling tactics (Meta structure)

• Budget scaling can reduce ROS
  • Injecting many new creatives increases Meta exploration and can temporarily dip ROS.
  • Recommendation: at a given budget, only inject the amount of creatives you expect to be confidently spent.
    • Example approach: “inject everything you’ll really spend,” then add later.
• Avoid excessive creative disruption (“too much AT” / too many rapid changes)
  • Meta optimizes at the account level, so big creative swings can impact the entire account.
• Testing structure preference
  • If CBO testing struggles to “catch up” on underperforming campaigns, test at:
    • the ad set level, or
    • via AB-style structure
  • Goal: give ads a better chance to decide early.

Whitelisting vs partnerships (ops + GTM)

• Whitelisting = page-level leverage
  • Use the same ad, swap whitelisted pages to access new audience pockets.
• Partnership = enabling creator/influencer connections inside the ad system
  • Treated like an MTA partnership-style setup.
• Action threshold KPI
  • If whitelisted partnerships < 30%, increase whitelisting to unlock higher spend capability and more segments.
• Concrete example
  • Place the same Winus ads on multiple whitelisted pages.
  • Only the page changes → improved spend because the page provides “leverage.”

Google Shopping listing playbook (“occupy more space”)

• Multiple product listings with different angles
  • Create several listings so searches like “foot massager” surface variants such as:
    • “foot massager for neuropathy”
    • “foot massager for well-being”
    • “foot massager for seniors”
  • Same product, different photo/angle → more banner real estate on Shopping.

AI ops initiative: “surgical AI” creative production (internal airlock)

A structured, quality-first AI workflow to generate high-volume static creatives without random output.

Stated objective

• Don’t create volume for volume’s sake.
• Produce hundreds of high-quality static ads using human review + AI processes.

The internal “airlock for creatives” (3 processes for now)

1. Copy / meaning adaptation from selected winners

   • Use an internal toolchain (referenced: TrainTrack payload, plus “SAS”) to pull creative inspiration from a curated set of brands/ads.
   • AI adapts messaging to the product mechanism.
     • Example: don’t only rewrite
       • “No more acne for mature skin” → “No more knee pain for old knees”
     • Instead, adapt based on the product’s unique mechanism, guided by a copywriter agent.
   • Human review gate before bulk generation.

2. Messaging testing via concept libraries

   • Build concept libraries containing winning concepts.
   • AI retrieves Trustpilot / competitor review signals and crafts message variants.
   • Test across multiple concept types, such as:
     • simple product image + line headline
     • before/after
     • premium black background content
     • drawings / cartoons

3. Hook → headline / LP iteration (static level)

   • For static ads, convert competitor winning hooks into headlines (described as a hack).
   • Possible future expansion: use scripts to generate additional LP / LP elements.

Additional AI operations notes

• Use Claude Code to publish ads without connecting directly to ad account control (time saver; lower ban risk).
• Build/optimize LP content carefully:
  • hero/long-form content shouldn’t be cut if it supports funnel conversion for colder audiences.

Product & offer strategy (differentiation under competition)

• Competition penetration tactic (awareness stage)
  • On landing pages (LP), show old version vs improved/new version.
  • Simple re-framing/renaming can make people perceive it as new → better, especially in competitive markets.
• Newness creates hope
  • When a product feels “new,” it attracts attention and increases conversion motivation before evidence fully accumulates.

Subscription vs one-time purchase (high-level execution guidance)

• If launching long-term with valuation/LTV goals and cashflow predictability:
  • Subscription recommended in ~80% of cases, especially with high-quality products to avoid high churn.
• For early experiments / fast cashflow:
  • Use non-recurring models to learn and launch faster.

Execution KPIs, targets, and constraints mentioned

• Ad “winner validation” thresholds
  • ROS above break-even; CPA under target CPA
  • Spend should exceed ~$5K to be considered truly winner-worthy
  • Performance should persist over time (“time” factor)
• Whitelisting mix target
  • Whitelisting < 30% ⇒ increase whitelisting
• Creative set sizing (Meta testing)
  • In testing (CBO/AB-type): 3–6 ads per batch/ad set
  • In CBO: up to ~15 ads per ad set (above that may be too much)
• Organic budget note
  • If budget is ~$8K, reserve ~20% (~$1.6K) for bio testing (AB-style organic-to-ads testing)
• Illustrative scaling claim
  • With ad winners at the account level, scaling from 10 → 30 creatives can reach > $1M/month, driven by account impact.

Customer targeting strategy

• Acquire new customers, not just retarget
  • Long-term winners should have:
    • low entry frequency (reach new audiences)
    • better prospects for scaling than purely high-ROS retargeting
• CTR / engagement timing preference
  • Prefer 1-day click behavior over 7-day click dominance for stronger immediate optimization signals.

Concrete examples & case cues

• Cartoon creatives
  • Referenced as producing strong comment signals and purchases.
• High spender anecdote
  • “$500k per day spend” LP tactic uses “old vs improved” renamed mechanism framing.
• TrainTrack ad library suggestion
  • Use external ad libraries for creative inspiration.

Sources / presenters mentioned

• Nico
• Matthéo (referenced leadership/operations via “Meto”)
• Sarah d’éné (quick iteration idea: voice changes targeting men/women)
• Marine and Lucas (mentioned in context of events/tools/setup)

Tools / brands referenced

• TrainTrack / TrendTrack
• Claude
• SAS (internal)
• TikTok Shop
• Meta (MTA)
• Trustpilot
• Amazon