3M€/MOIS, IA, GIFTING, META ADS, MARCHÉ US | MASTER #47

Company snapshot & performance metrics

• The brand is described as a US brand doing €3,005 last month (as stated in subtitles).
• Last year turnover/revenue was “a little over 100 million” (currency not clarified).
• Team size: ~12 people total (Martin + partner; others implied).
• Scaling lever: “small team + simplicity” (avoid complicating processes to scale).
• Repeat-customer strategy:
  • Keep it easy to repurchase via:
    • Best-seller since the beginning
    • Monthly variations (e.g., colors/variants)
  • Repeat customer rate is framed as one of the most important KPIs (no exact % given).

Growth “snowball” strategy (what scaled from end of last year to now)

• “Ecosystem snowballing” by continuously adding:
  • New creations (creative testing output)
  • New micro-influencers/ambassadors
  • More products/launches inside the same niche

• Data flywheel:

  • Customers + ambassadors + affiliate/micro-influencers generate data → the brand learns customers better → marketing/product cycles improve

• Research-to-creative pipeline:

  • They use market/creative intelligence tools to identify what angles/creatives work
  • Then push those angles into Meta (notably via Meta Ads)

Product strategy (portfolio design)

• Simple product approach:
  • Not a gadget; positioned as broad daily-use
  • Easier scaling due to clearer value proposition
• One flagship bestseller:
  • The flagship remains the main stability/growth driver
  • New products/variants support it rather than replacing it
• Variant cadence:
  • When in fashion category: “variations every month” (e.g., new colors)
• Launch volume (approx.):
  • New variants: ~300
  • Additional new products: ~20-something (exact count unclear)
• KPI emphasis:
  • Prioritize increasing repeat customers
  • Benefit: a stable bestseller reduces downturn risk

Marketing & sales engine: Meta Ads + gifting + whitelisting

Channel mix

• Primary: Meta Ads (“MTA” in subtitles), including work for recurring clients.
• Secondary channels:
  • Pinterest
  • Applovin
  • Google (described as the hardest for them)
  • Email exists but is largely supporting rather than the primary driver.

Email approach

• Pareto mindset for email (not heavy automation described).
• Emails operate in parallel, not as the core acquisition driver.

Influencer/ambassador “gifting” playbook (A to Z)

Inputs & selection

• Targeting:
  • Micro-influencers matched to an avatar/target audience
  • Followers can be as low as ~500 if content is strong.
• They also tested larger creators (100k–200k) and claim performance differences weren’t meaningful.

Sourcing & outreach operations

• Two full-time assistants handle:
  • Listing influencers
  • Outreach
• Sourcing method:
  • Manual/VA-driven search for less-contacted profiles (higher response rate)
• Mentions/discovery:
  • Keyword searches / Instagram-style discovery (tool specifics partly unclear)
• Platforms referenced (conceptually) include Modash (Europe) and Incense (US/worldwide), but they emphasize relying more on assistants/manual sourcing to avoid contacting the same creators everyone targets.

Deal structure (core differentiator = contracts + deliverables)

Step 1: Gifting

• Send 3–4 products per micro-influencer.
• Deliverable request:
  • 3 reels + 2–3 photos

Step 2: Contracts

• Influencer signs a contract specifying deliverables/obligations.
• Claimed result: ~90% return rate of gifted creators sending required videos.
• Contract includes distribution requirements:
  • Content must be posted (not only sent)
  • Enables whitelisting and ad reuse

Briefing method (freedom + winning angles)

• No heavy scripting.
• Provide:
  • Winning marketing angles
  • Hook examples
• Creator freedom:
  • Allowed execution variety to increase content diversity.
• Rationale:
  • Freedom without angles underperforms.
  • Over-scripting reduces diversity.
  • The “angles + hooks + freedom” blend is presented as critical.

Scale-up layer: “best ambassadors” with longer-term contracts

• After initial micro-influencers:
  • Top performers get long-term/month-based arrangements.
• Compensation model:
  • Either fixed minimum $1000
  • Or minimum $1000 + affiliate commission
  • They describe switching to “only affiliate marketing fee” if they exceed $1000.
• Production expectations:
  • Two reels per day, plus video sending and posting.

Enablement & coaching

• Coaching assets:
  • A Telegram channel with instructions and examples
  • Ambassadors are coached by assistants
  • A Discord-like structure and weekly meetings (ambassadors training each other)
• Goal:
  • Build an “army of people selling the brand” via incentives, coaching, and peer success sharing.

Whitelisting & incremental acquisition

• Ambassador content is reused in ads via Meta whitelisting.
• ROI is described as hard to isolate:
  • The impact is presented as ecosystem-wide (more brand presence → more data + reach).
• Claimed mechanism:
  • Ambassador postings expand audience reach beyond what the brand page could target directly
  • Increases the “catch pool” (example: Meta shows 300M potential buyers, but you won’t access that without leveraging the creator pool).

Creative testing & automation stack (AI usage)

Creative intelligence workflow (what triggers new tests)

• They monitor product traction (e.g., last 7 days).
• Identify top 10 creative aspects driving performance.
• Convert those findings into the angles to push in tests.

LIA + Claude usage (especially for statics)

• Claude/LIA used to generate ad creatives, especially statics.
• Simplified process:
  1. Collect data from post-purchase questionnaires/forms
  2. Produce a report → feed into “Cloud/Claude” (also referenced as “Cloud and Claude”)
  3. Claude outputs recommended angles/creation types → tested and iterated
• Reality check:
  • Automation requires human feedback
  • Wrong inputs → weak outputs
• They create loops:
  • generate → review → correct → regenerate

“Brute-force” creative philosophy

• Reuse what already worked:
  • Swap product + benefits on known-performing creative templates
  • Use competitor/big-brand top performing stats as reference (“don’t overthink it”)
• Best-performing visual format:
  • Simple facecam + product
  • Natural speaking, advice-style
• Overly complex or unnatural content performs worse.

Meta Ads account structure (measurement & scaling tactics)

• Keep it simple:
  • CBO per product
  • Testing like “by angle” or “by promo” didn’t perform as well.
• Testing philosophy:
  • Prefer fewer, cleaner structures to avoid fragmentation.
• “Zombie campaign / zombie CBO” concept:
  • Move “loser” ads/campaigns into a separate campaign
  • Keep paying/learning without corrupting core target CPA performance
• Cost-cap experiments:
  • Duplicate zombie strategy with Cost Cap
• CPA testing guidance:
  • If target breakeven purchase CPA is 30, test 30/35/40, then adjust (including 45/25) until target is found.

Product research tooling & “nugget” discovery trick

• Use Codata for product discovery.
• Workaround (“glitch”):
  • Use parameter filters with minimum turnover thresholds to hide products from users on classic/limited plans.
• Example:
  • If the limit is 100K, set minimum revenue ≥100K to surface emerging “nugget” products others can’t see easily.

Post-purchase questionnaire (critical data collection playbook)

Why it matters

• Described as one of the most important early implementations because LIA needs data.
• Captures:
  • Why customers bought
  • What was good vs. not good
  • Requests for what new variant/product to launch

KPI usage & closed-loop improvements

• They claim it can reveal:
  • The single ad that “blew everything up” last month (exact figures not provided)
• Feedback routing after purchase:
  • Score 1–10 (NPS-like)
  • If 1–6 → redirect to customer service
  • If 6–8 → redirect to site reviews
  • If 8–10 → redirect to “other review”/Trustpilot-like reviews
• Reported volume:
  • 2–3 Trustpilot reviews per day via this routing logic

Marketplace expansion: TikTok Shop + Meta “one-click checkout” (high level)

• TikTok Shop:
  • Planned; described as in progress
  • They didn’t prioritize it due to ongoing launches.
• Meta-side commerce:
  • Meta shop/webshop evolving with a PayPal one-click purchase partnership (described as “huge”).
• Amazon:
  • Opened Amazon and are sending inventory to Amazon warehouses
  • Expecting meaningful contribution
  • Industry reference: some brands do ~30% revenue on Amazon.

Org/process & recruiting insights (team scaling operations)

• Hiring reality:
  • Best hires come via recommendations/social networks, not LinkedIn ads.
  • Heavy use of Philippines support for roles (especially after-sales/assistants); managerial profiles also sourced there.
• Head-hunting approach:
  • Recruiters can miss fit when requirements are too specific.
  • More effective: manual senior profile search + trusted recommendations.
• Risk control:
  • When a recruiter recommends someone, quality is tied to recruiter reputation.
• “Team leverage” tactic:
  • Hire a trusted employee who recruits/train others (family/neighbor network described as an acceleration method).

Frameworks / playbooks explicitly referenced

• Ambassador “gifting + contract + whitelisting” playbook
  • Deliverables enforced via contract → ~90% return rate
  • Briefing blend: winning angles + hook examples + creative freedom
  • Scale top performers into long-term “army” model
• Creative testing intelligence loop
  • Product traction → identify top creative angles → test and push winning angles
• CBO-only Meta structure
  • Simplify structure to reduce account complexity
• Post-purchase questionnaire closed-loop system
  • Collect CX + motivations + improvement requests → feed AI for better creatives
  • Also route reviews/customer service based on score
• Codata “nugget” filter workaround
  • Use minimum turnover thresholds to find products limited-plan users can’t easily see

Key metrics/KPIs and targets mentioned

• Revenue/turnover
  • Last month: €3,005
  • Last year: >100M
• Return rate on gifted creators
  • ~90% (deliverable return rate attributed to contracts)
• Repeat customer KPI
  • “Most important currently” (no numeric target given)
• Creative production requirements
  • Micro-influencers: 3 reels + 2–3 photos
  • Top ambassadors: 2 reels/day (plus posting and video submission)
• Review volume
  • 2–3 Trustpilot reviews/day
• Meta testing targets
  • Example breakeven target CPA: 30
  • Test range guidance: 30/35/40, then adjust (including 45/25)
• Marketplace revenue benchmark
  • Amazon example: some brands do ~30% revenue on Amazon

Concrete recommendations / actionable takeaways

• Implement a post-purchase questionnaire immediately to:
  • Feed AI creative generation with real customer language
  • Improve ads using customer experience insight
  • Route customers to service vs reviews based on a happiness score
• Scale with an ambassador contract model:
  • Contracts + deliverables enforcement → dramatically improve content return rate
  • Require posting for whitelisting reuse in paid ads
• Brief creators with:
  • Winning angles + hook examples while keeping creative freedom for variety
• Keep Meta Ads structure simple:
  • Use CBO per product
  • Avoid over-segmentation by angle/promo if it underperforms
• Use intelligence-driven creative iteration:
  • Identify top creative angles from recent traction windows, then launch new tests quickly
• Recruiting:
  • Prefer recommendations/network and Philippines support for scalable operations
  • Be cautious with headhunter targeting to avoid misfit

Presenters / sources mentioned

• Martin Sur (main guest; e-commerce operator described)
• Lucas (presenter; co-discussed AI/creative processes)
• Matthéo (presenter; discussed recruiting and recruitment platforms)
• Nico (mentioned in context of ads/tools and Marketing Studio discussion)

Brand/tool names referenced

• LIA, Claude, Codata, FASMOS, Instantly, Cloud/C2 (as referenced), Trendi Track / Trend-Track (mentioned), Xfield (and “Marketing Studio”, “Scidence 2.0”)
• Commerce platforms: Meta / TikTok Shop / Amazon
• PayPal partnership (Meta commerce mention)
• Review/platform references: Trustpilot, Discord, Telegram
• Sourcing/research references: Modash, Incense, Upwork, online.ph, LinkedIn