Product → Mechanism → Angle

The single most important analytical move in this system. A product is sold through a mechanism (why the problem exists and why this fixes it). A mechanism opens a family of angles (the outcomes people want). Ads are angles. So the order of thinking is always:

PRODUCT  ──►  MECHANISM(S)  ──►  ANGLES (outcomes)  ──►  ADS
 what it is     the hidden cause     what the customer       one ad = one angle,
 what it does   + why this fixes it  wants, in their words   one sub-avatar, one promise

When writing ads the order flips: avatar → angle → (mechanism follows from the map). The map is built once, here; the ad writer never invents a mechanism on the fly.

---

1. Definitions (house meaning)

Rules for every mechanism:

• Three-link causal chain, never an ingredient alone: [ingredient] → [intermediate action] → [benefit the person FEELS]. Each link must be believable alone; if one needs a leap, add a link. Use explicit connectors ("that's why", "which means", "so").
• Simplify for a 14-year-old. No jargon to look expert. "Aromatase" may appear at most once on a landing page and never in an ad hook.
• Name it. A named mechanism is remembered, compared and asked for, and protects against clones.
• The UMS goes in almost every ad. The UMP can be dropped lower in the funnel (MOFU/BOFU).
• Never contradict an existing belief head-on. Start from it ("the ones you took were under-dosed"), then redefine the category against it.

---

2. Method: build the mechanism inventory for a new product

2.1 Collect candidate mechanisms (4 sources)

1. The product's own page and label: ingredients, dose, claims the brand is allowed to make.

2. Competitors' ads and landing pages (same product or same desire): every cause they accuse. Use the fixed list of "villains" seen in the testosterone market as a checklist for any category: hormonal conversion · stress hormone · nutrient deficiency of modern food · poor absorption / inflamed gut · endocrine disruptors / plastics · binding protein locking the hormone · blood flow / nitric oxide · precursors missing · natural decline with age · parasites / toxins · insulin / blood sugar · inflammation.

3. Voice of customer: what customers THINK is causing it ("I eat clean and still…", "my doctor said it's age"). Beliefs you can start from.

4. The "hidden mechanism" prompt (Evolve): "What does the product do that you don't advertise? What would customers complain about if it worked too well?" Plus: "Identify the new mechanisms that would resonate in [market]" with the product overview + URL loaded.

2.2 Fill a mechanism card for each candidate

Mechanism: [name] (enemy → fix)
- Enemy name said on air:
- Action name:
- Belief-bridge fact (accepted without proof) + source:
- Problem chain (3–5 links, plain words, source each):
- Solution in one sentence (A + B = C):
- Elimination triad (not X / not Y / not Z):
- Failed solutions it explains (name — why it misses this mechanism):
- Analogy a 14-year-old gets:
- Mechanism visual (3–5 step animation or static diagram):
- Forbidden terms + compliant rewrites:
- Freshness (how many shops scale on it?) / Belief (how plausible to the avatar?):
- Angles it can credibly reach:

2.3 Score Freshness × Belief (Crafter)

Check freshness on TrendTrack (how many advertisers scale on this cause in the last 60 days), Google Trends, viral TikToks, Reddit volume. Check belief against the avatar's existing beliefs (the bridge).

2.4 Pick a differentiation case (Crafter's three)

1. Format transplant — same desire + same mechanism, better delivery (gummy vs pill, capsule vs powder). Easiest and lowest risk: you inherit competitors' education. The main battle in every ad is why your format is better; mine the negative reviews of the other format.
2. Desire transplant — same mechanism, new desire (collagen skin → hair). Medium effort. Change the UMP/UMS wording and never mention the old desire's claim.
3. Core swap — new mechanism for the same desire. You pay all the education from zero.

2.5 Draw the map

One line per mechanism, then a matrix.

Estrogen conversion (fat converts T → estrogen → more fat) :: flat belly | lose the man boobs | shirt off at the pool | drive back | "she notices"
Stress hormone (chronic cortisol = storage mode)        :: belly that won't move | sleep through the night | stop cravings after meals | calm at 6 p.m.
Modern-food deficiency (ancestral organs fill the gap)    :: energy back by 10 a.m. | strength back at the gym | feel like myself again
Wild vs farmed source (potency of the source animal)      :: (comparison / anti-waste angles) "only one worked" | "stop paying for 12 pills"

---

3. Method: spin N angles from ONE mechanism

A validated mechanism is worth more than a validated ad, because it unlocks a family of angles. To spin angles:

1. Write the chain end to end, then list every downstream symptom the chain can honestly explain.
2. For each symptom, write the outcome the person wants (angle), the scene where they notice it, and who else notices.
3. Keep only symptoms that (a) the product can honour, (b) many people share, (c) are visible today (sell the repair, not the prevention).
4. Rank by observed market weight (competitor ads + VOC volume) and by distance from your current winning angle (closest first = cheapest test).

Worked example 1 — Oil of oregano (the founder's own example)

Mechanism chain: gut parasites → nutrients poorly absorbed → the body stores fat → fat makes estrogen.

Second mechanism for the same product: plastics/endocrine disruptors in water and food → the same man-boobs angle can be re-sold with a new cause (a Mechanism cross in Ad DNA = a new concept). Third: zinc deficiency for erections. That is: one product, three mechanisms, each with its own family of angles.

In-market confirmation (Resilia, radar study): oregano is their pillar product, sold through pre-landers on parasites, GLP-1, men 40+, prostate and gut reset; their two oldest winners ran 13 months on "Craving sweets all the time? Still bloated?" and "Still feeling off no matter how clean you eat?". One mechanism (gut), many doors.

Worked example 2 — Bison organ capsules (Wildhorn vs Primal Edge, same product)

Primal Edge's thesis carries everything: "A beer belly isn't a fat problem. It's an estrogen problem."

• UMP: an enzyme in belly fat converts testosterone into estrogen; estrogen orders more fat storage on belly and chest. A self-feeding loop. Soft conspiracy layer: "identified in 1979, doctors know".
• UMS (4 organs, 4 jobs): kidney = the blocker (selenium + zinc), liver = the regulator (vitamin A, B12, iron), heart = the burner (CoQ10 + carnitine), testicle = the source (androgen precursors).
• Anti-category reframe: "It's not a supplement problem. It's a FOOD problem." (14 carrots, 2 dozen oysters, 12 Brazil nuts, 9 ribeyes — or 2 capsules.)
• Angles that mechanism reaches: beer belly (king), man boobs / puffy nipples (strongest scroll-stop), drive in the bedroom, "feel like yourself again", shirt off at the pool.
• Separate mechanisms in the same market: cortisol belly (stress hormone; reaches cravings, sleep, puffy face, even dandruff at Primal Viking), modern-food deficiency (reaches energy, strength, "normal labs but not normal"), source potency (wild vs farmed: reaches comparison and anti-waste), GLP-1 depletion (reaches "lose fat, not strength").

Locked Wildhorn angle built from it: "Your fat is not fat." After 40 the body converts testosterone into estrogen → stored on belly and chest. Diets and the gym fight the wrong enemy. 4 jobs in one jar. Promise ladder: day 7 sleep, day 14 shirts, day 21 belt — or your money back.

Worked example 3 — Testosterone stack (Mars Men)

Mechanisms seen in their ads, with the angle each one carries:

• Endocrine disruptors ("plastics, seed oils, blue light — your grandfather had more T at 60") → generation angle, generic T boost.
• Cortisol → belly, energy, GLP-1 ("lose fat, not strength").
• Aromatase + SHBG ("produce, free, protect, optimize") → man boobs/belly, anti-TRT, bloodwork proof. Lesson: a stack product lets each ingredient carry its own mechanism ("tongkat blocks the enzyme that steals T, boron frees bound T, shilajit supports production") — "one ingredient = one job" is itself a proof device.

Worked example 4 — Cinnamon / blood sugar (Resilia)

Mechanism: blood sugar swings. Angle tested: "Support healthy blood sugar — all day long?" Sent straight to the product page, no advertorial; no winner after weeks. Read: a commodity mechanism with no named enemy and no pre-sell page = high sophistication, low freshness. Fix candidates: niche the sub-avatar (post-meal crash, prediabetic men worried about erections — Primal Storm moved hard to "prediabetes killed your erections"), or name a fresher enemy.

---

4. Method: read the mechanism out of ANY ad (review mode)

For each ad, extract in this order:

1. Symptom shown (what the viewer recognises).
2. Accused cause (the villain). Classify into the fixed list (§2.1). "None" is a valid answer — many promo ads have none.
3. Why prior solutions failed (diet, gym, pills, TRT, the doctor).
4. The fix (how the product acts on the cause).
5. The bridge (the story or authority that makes the jump believable: Vikings, a ranch, a Nigerian protocol, a founder, a study).
6. Angle (the outcome promised).

Then produce the spin-off table:

This is the format to use for any batch of videos or ads someone sends "to review": not a summary — an angle generator.

---

5. Pairing rule: the mechanism follows the angle

When an ad is being briefed:

1. The human (or the plan) picks the sub-avatar, then the angle.
2. Look the angle up in the map. The mechanism(s) listed against it are the only ones allowed.
3. If several mechanisms reach the angle, use the one closest to the "why" of the source ad being cloned.
4. If none reaches it: stop. Say "this product can't sell this angle" and propose the nearest angle that is on the map. Never improvise a link.

Between two angles on the same mechanism, only the language changes (the wanted result, the symptom, what was tried, the final promise). The cause and the fix stay identical. The copy chain (Ad DNA builder):

① result wanted, in the customer's words → ② symptom → ③ already tried → ④ hidden cause (UMP)
→ ⑤ what the product corrects (UMS) → ⑥ final promise that repeats ①

Prostate example: "sleep through the night" vs "sick of peeing all the time" — ①②③⑥ change, ④⑤ stay.

---

6. Mechanism quality checklist

• [ ] The enemy is named in 2–4 plain words with a physical verb (blocks, clears, locks, steals).
• [ ] The chain has 3–5 links, each believable alone, each sourced.
• [ ] It starts from a fact the avatar already believes.
• [ ] It explains why what they tried failed (permission: "it wasn't your fault").
• [ ] It makes the product the logical end of the chain (A + B = C).
• [ ] A 14-year-old can repeat it.
• [ ] It survives compliance: no disease claim, structure-function language ("supports", "helps maintain").
• [ ] Freshness × belief scored, differentiation case chosen.
• [ ] Angles listed, each with a scene and a witness.

---

7. Sophistication and the mechanism

Supplements in 2026 (sleep, testosterone, hair, blood sugar, menopause, libido, joints): assume level 3–4 by default. Niching resets the level: "weight loss" is 4–5, "arm fat after menopause" is lower, "…in women who already tried GLP-1" can be 1–2. The cheapest way out of a saturated market is not a new mechanism, it is a narrower sub-avatar. Escalate the mechanism, never the result number.

---

8. Output template (paste into the brand kit)

# [BRAND] — Mechanism map  (date, sources)

## Product
Physical: … | Functional: … | Price/offer: … | Allowed proof: … | Banned claims: …

## Map
M1 [name] (enemy → fix) :: angle | angle | angle
M2 …

## Mechanism cards
### M1 — [name]
(enemy, action, bridge fact + source, chain, A+B=C, triad, failed solutions, analogy, visual, banned terms,
 freshness/belief, angles)

## Angle families (ranked)
| Rank | Angle | Mechanism | Evidence (ads/days/LPs, VOC count) | Scene | Witness |

## Refused angles (no mechanism reaches them)
- [angle] — why refused