QUANTITATIVE CREATIVE STRATEGY ANALYSIS

Weight Loss — 127 Spy Creatives — February 2026

1. DATA CENSUS SUMMARY

Dataset Composition

All 127 creatives = Emagrecimento niche. No cross-niche data available. D3 (Problem) is essentially constant at WEIGHT_FAT (98.4%), and D9 (Transformation) is constant at AESTHETIC (97.6%), and D10 (Modality) constant at VERBAL_DOMINANT (98.4%). These three dimensions are invariant and won't discriminate between creatives — the strategic intelligence lives in the remaining 6 dimensions.

Top Products in Dataset

Frequency Distributions — The Discriminating Dimensions

Winners vs. All — Where the Alpha Lives

The most revealing comparison. Bold = statistically meaningful shift from population base rate:

Headline insight: Winners have MORE diverse avatar roles (less peer, more caregiver/expert), use MORE discovery architecture, lean on MORE authority-backed proof, and paradoxically use LOWER pain intensity. The market over-indexes on peer testimonials and pain amplification, but the winners are playing a more nuanced game.

2. CO-OCCURRENCE INTELLIGENCE

The 5 Strongest Co-occurrence Clusters

These combinations appear significantly more often than chance would predict (O/E > 1.5):

Cluster 1: "The Curiosity Investigator" (O/E 2.1-2.8)

QUESTION hook + DISCOVERER avatar + DISCOVERY_FIRST architecture + INFORMATION_GAP framing

When someone asks a question to open, they're almost always a Discoverer (O/E 2.79), they almost always use Discovery architecture (O/E 2.12), and Discoverers overwhelmingly pair with Information Gap (O/E 1.64) or Injustice (O/E 2.86). This is a coherent persuasive engine: curiosity opens the door, the discoverer role maintains it, and the revelation pays it off.

n = ~9-13 creatives consistently use this cluster

Cluster 2: "The Personal Journey" (O/E 1.5-2.4)

NARRATIVE hook + PEER avatar + TESTIMONIAL_FIRST architecture + GAIN_FRAMING

All 12 narrative hooks are Peer (O/E 1.49), and ALL 12 use Testimonial-First architecture (O/E 2.35). ZERO narrative hooks pair with Discovery or Demonstration. This is a locked combination — if you tell a story, it must be a personal journey with a positive outcome arc.

n = 12 creatives

Cluster 3: "The Action Demo" (O/E 1.8-1.9)

IMPERATIVE hook + DEMONSTRATION_FIRST architecture + GAIN_FRAMING

Imperatives strongly co-occur with Demonstration (O/E 1.85). Commands ("Watch this", "Look at this", "Stop scrolling") naturally lead to a visual proof element. Pairs with GAIN framing (positive transformation) rather than fear.

n = 11 creatives

Cluster 4: "The Authority Reveal" (O/E 1.7-2.8)

EXPERT/CAREGIVER avatar + DISCOVERY_FIRST architecture + AUTHORITY_STUDY proof

All 4 Experts use Discovery architecture (O/E 2.82). 5/6 Caregivers also use Discovery (O/E 2.35). Discovery architecture pairs strongly with Authority proof (O/E 1.66). This cluster uses credibility to license the revelation — "I'm qualified to tell you this hidden truth."

n = 8-9 creatives

Cluster 5: "The Gut Injustice" (O/E 2.9+)

GUT_MICROBIOME mechanism + INJUSTICE_ATTRIBUTION framing + DISCOVERER avatar

4 out of 6 gut microbiome creatives use Injustice framing — a wildly disproportionate co-occurrence. The gut angle naturally invites a villain narrative (Big Food, toxins, parasites). This is the "Metabolic Parasite" and "Ginger Trick" family.

n = 4 creatives (small but extremely coherent)

The 5 Most Notable Absences

These are theoretically valid combinations that NEVER appear or appear far less than expected:

3. ARCHETYPE CATALOG

Archetype A: "The Peer Believer" — The Market Default

Match count: ~35-40 creatives (28-31% of dataset) Products: Gelatin Trick, Pink Salt, Chocolate Trick, OAT Trick Persuasive logic: Minimum cognitive friction. "I'm like you, I was struggling, I found this thing, look at what happened." Social proof + visual proof of results. Solution-dominant (MSOL) means most time explaining the product, not the problem. Ceiling: This is the most saturated archetype. Viewer fatigue from seeing dozens of identical peer testimonials. Every competitor runs this. It works but has diminishing marginal returns. Winner rate: ~30% of this archetype reaches Escalada — BELOW the dataset average of 34%.

Archetype B: "The Curious Revealer" — The Discovery Engine

Match count: ~15-18 creatives (12-14%) Products: $1 Pink Gelatin, Pink Salt Hack, Bariatric Jelly, Metabolic Parasite Persuasive logic: Opens with a "secret" or "discovery" that creates an information gap. Discoverer role maintains credibility as an insider/whistleblower. Authority study pays off the curiosity. The viewer feels they're getting access to suppressed information. Ceiling: Depends on novelty of the "secret." When the same secret has been revealed in 50 other ads, the information gap collapses. Requires constant angle rotation. Winner rate: ~45% reach Escalada — ABOVE average. Discovery-first over-indexes in winners.

Archetype C: "The Visual Prover" — Show Don't Tell

Match count: ~15-18 creatives (12-14%) Products: Pink Salt, Bariatric Gelatin, OAT Trick Persuasive logic: "Look at this" / "Watch what happens" — commands stop the scroll, then immediate visual proof anchors belief. Most effective when the demonstration is genuinely surprising or novel. Ceiling: Requires strong visual assets. Hard to differentiate when competitors have similar before/afters. The demonstration needs to be genuinely novel to break through. Winner rate: ~40% reach Escalada — slightly above average.

Archetype D: "The Authority Insider" — Trust Through Credentials

Match count: ~10-12 creatives (8-9%) Products: $1 Pink Gelatin, Gelatin Trick, Bariatric Gelatin Persuasive logic: Credentialed speaker (doctor, nurse, concerned family member) reveals a discovery. The authority proof is pre-licensed by the speaker's role. This creates a dual trust signal: "This person is qualified AND the science backs it up." Ceiling: Requires believable expert/caregiver actors. Compliance risk with health claims from "doctors." BUT: this archetype has the highest win rate in the dataset — Expert at ~80% win rate, Caregiver at ~71%. It is massively UNDER-explored relative to its performance.

Archetype E: "The Gut Rebel" — The Injustice Engine

Match count: ~4-6 creatives (3-5%) Products: Ginger Trick, Metabolic Parasite Persuasive logic: "You've been lied to. The real reason you can't lose weight is a gut parasite/microbiome issue that the diet industry doesn't want you to know about." Anger/outrage drives engagement and shares. Problem-dominant (MUP) builds tension before offering the solution. Ceiling: Small sample size. Compliance and platform risk with injustice/conspiracy framing. But when it works, it creates deeply engaged audiences. The only archetype that occasionally promises non-aesthetic transformation (identity/vitality).

4. WHITE SPACE MAP

Tier 1: Completely Absent (0 examples)

Tier 2: Under-Explored (1-5 examples, but promising signals)

5. THE 10-CONCEPT THESIS

Portfolio Construction

Portfolio floor: Even if all HIGH-risk and half of MEDIUM fail, we validate 4-6 concepts = sufficient creative fuel for scaling.

LOW RISK CONCEPTS (4)

Concept 1: "The Peer Discovery Shift"

Target DNA:

Strategic rationale: Takes the most common avatar (Peer, 67%) but shifts from the over-saturated TESTIMONIAL_FIRST to DISCOVERY_FIRST architecture, which over-indexes in winners (+7pp). Adds AUTHORITY_STUDY proof (winner-preferred) instead of testimonial proof. Essentially the "Peer Believer" upgraded with the Discovery Engine's structure.

Risk level: LOW — Both components are proven individually. The combination is common but this specific configuration optimizes for winner-correlated dimensions.

4 Hook variations:

1. DECLARATIVE: "I was 52 and had tried everything until a friend showed me what doctors are now calling the #1 metabolism discovery of 2026."
2. IMPERATIVE: "Stop blaming your willpower — here's the metabolic switch that a Stanford study just confirmed actually controls fat storage after 50."
3. QUESTION: "Did you know that a simple discovery is helping thousands of women over 50 burn stubborn fat without dieting or exercise?"
4. NARRATIVE: "Last March, I stumbled on something that changed everything I thought I knew about weight loss after menopause."

Concept 2: "The Caregiver's Revelation"

Target DNA:

Strategic rationale: Caregiver avatar is 2x over-represented in winners (12% vs 6% overall) — the single strongest avatar signal in the data. Combined with DISCOVERY_FIRST (winner-preferred) and CONDITIONAL hook (strong audience filter that creates instant relevance). The "my mother/wife tried this" frame reduces skepticism because it doesn't feel like a first-person pitch.

Risk level: LOW — Caregiver + Discovery is a proven co-occurrence (O/E 2.35). Just formalizing a winning pattern that exists in small numbers.

4 Hook variations:

1. CONDITIONAL: "If someone you love is struggling to lose weight after 50, this 30-second discovery could change their life."
2. DECLARATIVE: "My mother lost 34 pounds at 63, and it wasn't from a diet or the gym — it was this one thing her doctor finally admitted."
3. QUESTION: "What would you do if you found the one thing that could finally help your wife lose the weight she's been fighting for years?"
4. IMPERATIVE: "Watch what happened to my mom after she started doing this one simple thing every morning."

Concept 3: "The Expert Endorsement"

Target DNA:

Strategic rationale: Expert avatar has an ~80% win rate (4/5 are winners). Experts naturally pair with DISCOVERY_FIRST (O/E 2.82) and AUTHORITY_STUDY proof. GAIN framing matches the Expert+Gain co-occurrence (O/E 1.56). HORMONAL mechanism gives the expert something credible to explain. This is not experimental — it's formalizing the single highest-performing avatar and building the optimal shell around it.

Risk level: LOW — Every component is proven. Just scaling volume on a proven but under-explored pattern.

4 Hook variations:

1. DECLARATIVE: "As an endocrinologist, I've helped over 3,000 women lose weight — and it starts with understanding one hormone that controls everything."
2. QUESTION: "Why do 90% of diets fail for women over 50? As a hormone specialist, I can tell you it has nothing to do with willpower."
3. IMPERATIVE: "Listen carefully if you're a woman over 50 who can't lose weight — what I'm about to share contradicts everything you've been told."
4. NARRATIVE: "In 22 years of treating women for hormone-related weight gain, I've never seen anything work as consistently as this."

Concept 4: "The Demo Proof Machine"

Target DNA:

Strategic rationale: Imperative + Demonstration is a strong co-occurrence (O/E 1.85). Uses 1_MILD intensity (winner-preferred, 16% vs 9% base rate — counter-intuitive but data-backed). The understated "it was so easy" tone + visual proof = powerful combo. Before/After proof complements the demonstration architecture naturally.

Risk level: LOW — Proven structure. The twist is using mild intensity (which the data says winners prefer) instead of the default moderate.

4 Hook variations:

1. IMPERATIVE: "Watch what this pink powder does in just 14 days — the results speak for themselves."
2. DECLARATIVE: "This simple morning routine helped me drop 2 dress sizes without feeling like I was dieting at all."
3. QUESTION: "Want to see what losing 23 pounds actually looks like on a real woman over 55?"
4. CONDITIONAL: "If you've been wondering whether those before-and-after photos are real — let me show you mine, step by step."

MEDIUM RISK CONCEPTS (4)

Concept 5: "The Gut Whistleblower"

Target DNA:

Strategic rationale: The GUT + INJUSTICE co-occurrence is the most intense signal in the dataset (4/6 gut creatives use injustice). DISCOVERER + INJUSTICE (O/E 2.86) and QUESTION + DISCOVERER (O/E 2.79) are the two strongest pairwise signals. This concept stacks THREE of the strongest co-occurrence signals. Problem-dominant (MUP) builds tension before the solution.

Risk level: MEDIUM — Each pairwise signal is proven, but the full stack is rare (n=4). Injustice framing carries compliance risk. The gut-weight connection needs careful scripting to feel credible, not conspiratorial.

4 Hook variations:

1. QUESTION: "Is a tiny gut parasite the reason you can't lose weight no matter what you try?"
2. IMPERATIVE: "Stop counting calories — a gut biologist just revealed why your microbiome has been sabotaging your weight loss for years."
3. DECLARATIVE: "The $72 billion diet industry doesn't want you to know that weight gain after 50 starts in your gut, not on your plate."
4. CONDITIONAL: "If you've tried every diet and exercise program without lasting results, the problem might be microscopic — and it's living in your gut right now."

Concept 6: "The Future Self"

Target DNA:

Strategic rationale: Future pacing (D9b=YES) appears in 16% of winners vs 10% overall — a 1.6x over-representation that suggests untapped potential. This concept takes the proven "Personal Journey" archetype (NARRATIVE + PEER + TESTIMONIAL) and amplifies the ending with explicit future-self visualization. "Imagine walking into your daughter's wedding feeling confident..." The hormonal mechanism gives it a science backbone.

Risk level: MEDIUM — Base structure is proven (Archetype A/B). The bet is that explicitly adding future pacing lifts the persuasion ceiling. The data hints this is true but the sample is small.

4 Hook variations:

1. NARRATIVE: "Six months ago, I couldn't look at myself in the mirror. Today, I'm shopping for a dress for my daughter's wedding — and I actually feel excited."
2. DECLARATIVE: "At 58, I never thought I'd feel this confident again — but here I am, 31 pounds lighter and imagining a future I'd given up on."
3. QUESTION: "Can you picture yourself 90 days from now, stepping on the scale and finally seeing the number you've been dreaming about?"
4. IMPERATIVE: "Close your eyes for a second and imagine waking up tomorrow with a flat stomach, boundless energy, and the confidence to wear whatever you want."

Concept 7: "The Expert Investigator"

Target DNA:

Strategic rationale: This fills the EXPERT + INFORMATION_GAP white space (0 current examples). Both Expert (80% win rate) and Information Gap (42% of winners) are strong individually, but they've never been combined. The hypothesis: an expert who reveals hidden information is more credible than a discoverer doing the same. The expert's credentials pre-license the revelation, making the information gap feel less conspiratorial and more "insider knowledge."

Risk level: MEDIUM — Strong theoretical foundation. Combines two winner-correlated elements that have simply never been put together. The risk is that experts and info-gap may feel contradictory (experts are transparent, info-gap implies secrecy).

4 Hook variations:

1. QUESTION: "Why won't your doctor tell you about the metabolic switch that controls weight gain after 50? As a nutritional scientist, I'm going to."
2. DECLARATIVE: "There's a metabolism discovery that Big Pharma doesn't want mainstream — and I have 20 years of research to prove it works."
3. IMPERATIVE: "Pay attention if you're over 50 — what I'm about to reveal has been suppressed by the diet industry for over a decade."
4. CONDITIONAL: "If you're a woman over 50 who feels like her metabolism has shut down, there's something the medical establishment hasn't been telling you."

Concept 8: "The Conditional Filter"

Target DNA:

Strategic rationale: CONDITIONAL hooks are the rarest (4%) but they create the strongest audience self-selection. "If you're a woman over 50 who..." instantly creates relevance for the target and screens out everyone else, potentially improving CTR quality and downstream metrics. Combined with Caregiver (winner-correlated) and a Gain frame. This tests whether the audience-filtering effect of conditional hooks improves not just CTR but conversion.

Risk level: MEDIUM — Conditional hooks are rare in the dataset but not absent. The bet is that the quality of the attention (highly self-selected) compensates for potentially lower volume. Combined with the caregiver's empathetic frame, this could perform very differently from the declarative-peer default.

4 Hook variations:

1. CONDITIONAL: "If you're a woman over 50 who's tried everything to lose belly fat, my mother's story might change everything for you."
2. CONDITIONAL: "For women who feel like their metabolism stopped working after menopause — there's a reason, and there's a fix."
3. CONDITIONAL: "If someone you love is struggling with stubborn weight gain after 50, you need to see what happened when my wife tried this."
4. CONDITIONAL: "When nothing else has worked — not the diets, not the gym, not the willpower — this is what my mother found instead."

HIGH RISK CONCEPTS (2)

Concept 9: "The Loss Amplifier"

Target DNA:

Strategic rationale: This concept violates EVERY market convention simultaneously:

• PROBLEM_FIRST architecture (0/127 use it)
• LOSS_FRAMING (2/127 use it)
• 3_SEVERE intensity (only 5% of winners use it)
• MUP_DOMINANT emphasis (problem-cause-focused)

But the theoretical case is strong: Kahneman's Prospect Theory shows loss aversion is 2x stronger than gain motivation. The entire market is gain-framed — if loss framing works even moderately, it would be maximally differentiated. Opens with the COST of inaction ("Every day you wait, your metabolism slows another 2%"), builds the problem cause, then pivots to the solution with before/after proof.

Risk level: HIGH — This is a contrarian bet. The market's absence of loss framing could mean (a) it doesn't work on Meta, (b) it creates compliance issues, or (c) nobody's tried it. We're betting on (c). If it validates, it unlocks an entirely new creative direction.

4 Hook variations:

1. DECLARATIVE: "Every year after 50, your body destroys an additional 5% of its metabolic capacity — and most women don't realize it until it's too late."
2. QUESTION: "Do you know what's happening to your metabolism right now, while you watch this? It's slowing down — and it won't stop on its own."
3. IMPERATIVE: "Stop ignoring this — the weight you gained this year won't be the last. Here's why it gets worse every single month."
4. CONDITIONAL: "If you're over 50 and you've noticed the weight getting harder to lose each year, what I'm about to tell you explains exactly why — and why waiting will only make it worse."

Concept 10: "The Identity Restoration"

Target DNA:

Strategic rationale: D9 = AESTHETIC dominates at 97.6%. Literally every ad promises "you'll look thinner." This concept tests whether an alternative transformation promise — IDENTITY_VITALITY ("I feel like myself again", "I got my life back", "I recognized myself in the mirror for the first time in years") — can outperform the aesthetic default for the 50+ audience. The hypothesis: for older women, weight loss is often a proxy for identity loss (feeling invisible, losing confidence, social withdrawal). Speaking to the REAL desire, not the surface desire, could create deeper emotional resonance and stronger conversion.

Risk level: HIGH — Only 1/127 creatives use non-aesthetic transformation. This could mean the market knows aesthetic works best, or it could mean nobody has tested the alternative. The narrative+peer structure is proven (Archetype A), so the ONLY variable being tested is the transformation promise. Clean A/B testable.

4 Hook variations:

1. NARRATIVE: "I used to avoid mirrors. I used to cancel plans. I used to feel like the real me had disappeared. Then, at 57, something changed."
2. DECLARATIVE: "Losing 28 pounds was great. But getting my confidence back? Getting my LIFE back? That's the part nobody prepared me for."
3. QUESTION: "When was the last time you felt like yourself? Not thinner — YOURSELF. The woman who loved getting dressed up, who said yes to every invitation?"
4. IMPERATIVE: "Forget about the number on the scale for a minute. Let me tell you what it actually feels like to wake up and recognize yourself again."

6. EXECUTION PRIORITY MATRIX

If We Can Only Test 5, These 5 — and Why

Why this ordering?

• C3 and C2 are near-certain to validate based on data signals — they establish our floor
• C5 has the most explosive upside (injustice framing creates viral sharing + deep engagement)
• C7 is the cleanest white-space test — if Expert + Info Gap validates, we can apply it across all products
• C10 is the biggest potential paradigm shift — if identity > aesthetic for this audience, every future concept changes

Concepts 6-10 ordered:

6. C1: Peer Discovery Shift — solid but less differentiated, can be added later
7. C4: Demo Proof Machine — proven structure, save for scaling phase
8. C6: Future Self — interesting but incremental on existing archetype
9. C8: Conditional Filter — good test but lower strategic impact
10. C9: Loss Amplifier — highest-risk, save until we have creative bandwidth to absorb potential failure

Key Production Notes

1. Each concept gets 4 hooks = 40 total hooks across 10 concepts. At the validation stage, each hook is a separate ad (different first 5 seconds, same body).
2. The body script can be templated per concept — the DNA signature defines the structure. Only the hook (D1) varies per ad within a concept.
3. Winners will be identifiable within 48-72 hours at $20-50/hook spend. Look for CTR > 2% and thumb-stop rate > 35% as primary validation signals.
4. Concepts that validate at hook level should immediately get body script variations — testing D5 architecture and D6 proof variations within the winning concept DNA.

Analysis generated: February 2026 Data source: 127 analyzed spy creatives from Supabase spy_creatives table Methodology: Frequency distributions, Observed/Expected co-occurrence ratios, archetype clustering, white space mapping