JungleLab learnings

Universal · 4 learnings

• universal A starved losing arm is a rejection by the auction, not a rival that was never measured. Never re-run against a rejected arm; the delivered winner already settled that contest.
  TBF Exp 16 — Scs lost discovery bridge · Jul 13, 2026
• universal Seedance Mini drop-in path is structurally blocked for character-sheet pipelines — BytePlus moderation rejects cartoon character references as 'real person.' Use keyframe-first arms (gpt-image-2 then video model) for any Seedance path involving character sheets.
  TBAB Exp 4 — Pixar scene renderer cost decomposition · Jul 10, 2026
• universal A starved losing arm is a rejection by the auction, not a rival that was never measured. Meta preferred the faster cut; that preference is the verdict.
  TBF Exp 9 — Joan Ginther VSL playback speed test · Jul 2, 2026
• universal Do not over-read the stale 90-day master table when the iteration dashboard shows a newer winner. For active/recent ITER ad sets, current Meta delivery can flip after the master snapshot.
  TBF Exp 12 — Ted talk hook framing cursed jetsetters · Jun 17, 2026

Style · 10 learnings

• style: Customer UGC Do not clean up harmless spoken disfluencies in AI talking-head testimonials. When Marge said 'six six' instead of 'sixty-six' and stumbled on 'Saint Casimir's,' those imperfections were load-bearing — the platform directed nearly all delivery to the disfluent original while ignoring the clean-pronunciation twin. The imperfection IS the authenticity.
  TBF Exp 34 — Marge spoken disfluency · Jul 26, 2026
• style: Customer UGC When re-rendering or iterating on an AI testimonial that won, preserve any natural-seeming stumbles or pronunciation quirks. A perfectly clean delivery reads as synthetic. The winning Marge had two minor disfluencies; removing them made her uncanny enough that the platform ignored the variant outright.
  TBF Exp 34 — Marge spoken disfluency · Jul 26, 2026
• style: Customer UGC For future AI testimonials, intentionally introduce one or two controlled, harmless disfluencies (a slight stumble, a self-correction, a natural pause-filler) rather than scripting a flawless delivery. The imperfection signals real person to the platform.
  TBF Exp 34 — Marge spoken disfluency · Jul 26, 2026
• style: raw_ugc_montage Use GPT 5.6 Sol for silent montage storyboards — it produces scene-1 pain-proof shots and correct product identity on the first attempt, where DeepSeek V4 Pro invents wrong products and writes forbidden text into visual fields.
  TBF Exp 22 — Storyboard model bakeoff raw ugc montage · Jul 19, 2026
• style: raw_ugc_montage When storyboarding a silent montage (no voiceover), the storyboard IS the persuasion. Don't treat the storyboard model as a commodity — the model that writes the scene beats and keyframe prompts determines whether the cold-feed scroller keeps watching.
  TBF Exp 22 — Storyboard model bakeoff raw ugc montage · Jul 19, 2026
• style: raw_ugc_montage Kimi K3 wrote a valid storyboard on first attempt but Meta didn't fund it — keep it as the fallback if GPT 5.6 Sol availability changes, since it passed the offline gate where DeepSeek failed.
  TBF Exp 22 — Storyboard model bakeoff raw ugc montage · Jul 19, 2026
• style: ugc_talking_head Don't flip the Hedra resolution default yet — wait for at least two delivering pairs across the three replication adsets before calling resolution load-bearing vs. noise.
  TBF Exp 15 — Vsl delivery resolution 540p · Jul 10, 2026
• style: Customer UGC Disproven — using a real actor doesn't improve results of UGC video.
  TBF Exp 1 — Real actor vs AI for UGC (Hassan) · May 11, 2026
• style: Customer UGC The script/concept matters more than whether the talent is real or AI.
  TBF Exp 1 — Real actor vs AI for UGC (Hassan) · May 11, 2026
• style: Customer UGC No need to invest in real actor shoots for UGC-style creatives when AI avatars perform equivalently.
  TBF Exp 1 — Real actor vs AI for UGC (Hassan) · May 11, 2026

Product · 6 learnings

• product Mechanisms can be rotated to fight ad fatigue without losing performance.
  JA Exp 4 — Mystical mechanism family test · Jun 12, 2026
• product New canonical frame: treat Energy Flow as one rotation in a mystical-cause family. Do not hard-lock it as the brand mechanism.
  JA Exp 4 — Mystical mechanism family test · Jun 12, 2026
• product Transferable principle: Replace age fatalism with a specific, felt alternative cause. Mystical/energy framing > clinical framing when both carry the reversal.
  JA Exp 3 — Static ad age-fatalism reversal test · Jun 6, 2026
• product The JA evolver should treat "Energy Flow + not age" as the canonical static ad frame.
  JA Exp 3 — Static ad age-fatalism reversal test · Jun 6, 2026
• product The JA evolver should use Energy Flow as the primary ad-congruence signal for page hero/story alignment.
  JA Exp 2 — Energy Flow big idea (current winner)
• product Landing-page experiments should treat the King of Medicine frame as the strongest current ad-market signal unless newer evidence supersedes it.
  JA Exp 1 — big idea angle test

Avatar · 86 learnings

The Forgotten Knowledge Collector · JA · 10 learnings

• avatar The Forgotten Knowledge Collector accepts the two-pathway king when it is shown, not lectured — Exp 013's graft failed on presentation, not mechanism, so keep the mechanism and drop the scholarly register.
  JA Exp 14 — FKC UMS vivid-presentation isolation on the Redmond skeleton · Aug 28, 2026
• avatar The Forgotten Knowledge Collector keeps the heritage 'restore the forgotten king' explanation at the plain-language register; a mechanism that works in the literate register must be re-expressed in the reader's own idiom, not transplanted as-is.
  JA Exp 13 — FKC champion assembly via UMS graft · Aug 15, 2026
• avatar A present-day shelf case in the omniscient voice is not a proven substitute for a sourced history for the Forgotten Knowledge Collector; treat the historical spine as still required.
  JA Exp 12 — FKC omniscient voice vs historical subject (the confound in Exp 007) · Aug 8, 2026
• avatar The Forgotten Knowledge Collector rewards simpler sentences and shorter words on the same beats. Do not treat the proven skeleton's reading grade as load-bearing.
  JA Exp 11 — FKC doc-VSL reading-level isolation · Aug 3, 2026
• avatar When scaling a Lane-A big-idea winner for the Forgotten Knowledge Collector, transplant only its opener onto the proven documentary skeleton — a full front-half rebuild earns nothing the simple swap does not already capture.
  JA Exp 10 — FKC turmeric-verdict on the scale surface (construction batch) · Aug 3, 2026
• avatar Do not default to elaborate front-half reconstructions (Decade of Turmeric, shelf-walk, time-gradient, skeptic's audit) — the collector already has a scale format and a new idea only needs its first beats.
  JA Exp 10 — FKC turmeric-verdict on the scale surface (construction batch) · Aug 3, 2026
• avatar Name the ingredient path in the solution span; the Forgotten Knowledge Collector responds to a mechanism you can point at, not a restoration myth.
  JA Exp 9 — FKC UMS isolation test: which Unique Mechanism of the Solution explains Joint Ascent best · Aug 2, 2026
• avatar The mainstream-shelf comparison frame ('I gave turmeric its whole decade') is the second viable big-idea family for the Forgotten Knowledge Collector; open FKC scripts on supplement fatigue as readily as on the drawer-swap.
  JA Exp 6 — FKC Big-Idea Concept Batch (video UGC) · Jul 21, 2026
• avatar Drawer-swap and turmeric-comparison compete for the same audience slot; the auction backs one and starves the other, so build them in separate rounds rather than one ad set.
  JA Exp 6 — FKC Big-Idea Concept Batch (video UGC) · Jul 21, 2026
• avatar Named authority (Haritaki), outcome-first with no collector persona, first-discovery pride and verification-story frames were not funded; do not default new FKC concepts to them.
  JA Exp 6 — FKC Big-Idea Concept Batch (video UGC) · Jul 21, 2026

The Overwhelmed Professional · KB · 10 learnings

• avatar Don't name-drop Ozempic or Mounjaro in KB statics for the Overwhelmed Professional — the food-noise concept alone carries enough recognition. The named-brand variant never cleared the delivery threshold, while the no-brand version got all the traffic. The medical-product names add policy risk without adding audience pull.
  KB Exp 6 — Glp 1 brand salience test · Jul 15, 2026
• avatar When writing KB statics that compete near GLP-1 awareness, lead with the felt experience ('Food noise is everywhere') rather than the cultural reference. The audience already lives in that world — they don't need you to name the drug to know what 'food noise' means.
  KB Exp 6 — Glp 1 brand salience test · Jul 15, 2026
• avatar: Overwhelmed Professional Rhetorical questions about whether stress-free life is possible don't create enough urgency — the overwhelmed professional already knows it's possible for other people; the question doesn't land
  KB Exp 4 — Stress hook type test · Jun 11, 2026
• avatar: Overwhelmed Professional Nostalgic memory hooks (remember when you felt at peace?) may feel too distant and retrospective — this avatar is in the thick of it, not looking back
  KB Exp 4 — Stress hook type test · Jun 11, 2026
• avatar: Overwhelmed Professional Physical tension is the strongest hook for the overwhelmed professional — it's concrete, felt in the body, and instantly recognizable
  KB Exp 1 — Pain point test
• avatar: Overwhelmed Professional This avatar responds to symptom-first copy over feeling-first or emotional hooks
  KB Exp 1 — Pain point test
• avatar: Overwhelmed Professional Body-first framing outperformed all other pain point angles
  KB Exp 1 — Pain point test
• avatar: Overwhelmed Professional "Noise" and "calm" resonate but less than physical tension — feeling-first works, just not as well as symptom-first
  KB Exp 1 — Pain point test
• avatar: Overwhelmed Professional Sleep disruption is relatable but low-urgency compared to body tension — this avatar may not frame their problem as a sleep issue
  KB Exp 1 — Pain point test
• avatar: Overwhelmed Professional Vague emotional framing ("life feels heavier") is too indirect — they need a concrete symptom, not a feeling
  KB Exp 1 — Pain point test

The Perimenopause Skeptic · KB · 5 learnings

• avatar For the Perimenopause Skeptic, keep the Historical Documentary VSL on the product-origin thesis — the story of monks discovering the Spirit Gate. Medical-dismissal and 3AM-false-alarm framings never earned delivery; the origin story took all the traffic. This avatar doesn't want her wound named in the opener; she wants the product's history.
  KB Exp 8 — Perimenopause big idea · Aug 1, 2026
• avatar When writing the next Perimenopause Skeptic script batch, develop execution variants within the origin-story frame rather than trying new thesis territory. Vary the opening hook, the pacing, or the visual treatment — but don't change the big idea. The frame works.
  KB Exp 8 — Perimenopause big idea · Aug 1, 2026
• avatar: Perimenopause Skeptic The named-skeptic + rational-worldview + one-question respect structure earns attention from the Perimenopause Skeptic. She listens when a credible named scientist asks the one question her rational worldview cannot answer.
  KB Exp 7 — Skeptic respect translate · Jul 20, 2026
• avatar: Perimenopause Skeptic But the respect opener captures attention without converting to purchase at meaningful volume. The biological-reframe territory from KB Exp 004 carries well once the opener hooks, but the distance from credibility to offer handoff is too long.
  KB Exp 7 — Skeptic respect translate · Jul 20, 2026
• avatar: Perimenopause Skeptic Keep the respect structure as a KB skeptic opener candidate, but compress the body. Shorten the biological-reframe section so the mechanism arrives faster after the respect hook.
  KB Exp 7 — Skeptic respect translate · Jul 20, 2026

The Fortune Seeker · LLM · 6 learnings

• avatar Disproven — LOW emotional state statics do not lead to high conversions for LLM Fortune Seeker avatar.
  LLM Exp 2 — Egyptian theme static ads - LOW emotional state · May 15, 2026
• avatar The Egyptian theme resonates with the audience, but emotional state framing matters — low/melancholic tone underperforms.
  LLM Exp 2 — Egyptian theme static ads - LOW emotional state · May 15, 2026
• avatar Static ads for this avatar should lean into higher-energy or aspirational emotional states rather than subdued tones.
  LLM Exp 2 — Egyptian theme static ads - LOW emotional state · May 15, 2026
• avatar Disproven — Kira-collab-driven AI copywriting given a proven avatar is not effective by itself.
  LLM Exp 1 — Kira-collab AI copywriting for proven avatar (Gary Henderson) · May 14, 2026
• avatar A proven avatar/backstory alone is not sufficient input for AI to generate performant ad scripts.
  LLM Exp 1 — Kira-collab AI copywriting for proven avatar (Gary Henderson) · May 14, 2026
• avatar Human creative direction and iteration are still required; the AI collab process needs stronger constraints or examples.
  LLM Exp 1 — Kira-collab AI copywriting for proven avatar (Gary Henderson) · May 14, 2026

The Healer · LLM · 1 learnings

• avatar: Healer Use a low, serious American documentary/podcast register for Healer VSLs. The theatrical British documentary voice and the earlier Chinese-American medical-authority pick both lost once the auction matured.
  LLM Exp 3 — Realistic 3x4 narrator voice · Jul 10, 2026

The Dimmed · TBAB · 4 learnings

• avatar: Dimmed Don't position TBAB as something "for elites" when talking to The Dimmed — they don't see themselves as hedge fund managers or surgeons, so the product feels like it's not for them
  TBAB Exp 2 — Post hook framing test · Apr 29, 2026
• avatar: Dimmed Ground the post-hook in a specific felt moment ("3am brain solving things") rather than an identity claim ("top performers use this") — The Dimmed responds to "I've felt that" not "I want to be that"
  TBAB Exp 2 — Post hook framing test · Apr 29, 2026
• avatar: Dimmed "Forbidden knowledge" hooks ("the elders didn't want me to share this") stop thumbs but attract the wrong people — curious browsers, not buyers
  TBAB Exp 1 — Hook framing test · Apr 24, 2026
• avatar: Dimmed The Dimmed might need a hook that's as attention-grabbing as secrecy but attracts people who actually want to fix their brain fog — that hook hasn't been found yet in this batch
  TBAB Exp 1 — Hook framing test · Apr 24, 2026

The Spiritual Seeker · TBAB · 1 learnings

• avatar Open Spiritual Seeker primary text on the Stanford proof, written as short standalone beats in the Overlooked one-liner rhythm.
  TBAB Exp 5 — Primary text entry thesis · Jul 25, 2026

17 Jetsetters · TBF · 1 learnings · not in avatar registry

• avatar Keep envy and status-jealousy phrases as the named social-threat for 17 Jetsetters. Comparison and dismissal lost.
  TBF Exp 25 — Social threat phrase type · Jul 21, 2026

The Ageing · TBF · 1 learnings

• avatar For The Ageing documentary VSL, a closing-passage cleanup alone is not a lever; the next iteration should change the selling block, not polish the leak.
  TBF Exp 60 — Ageing writer context repair · Aug 27, 2026

The Ailing · TBF · 5 learnings

• avatar Don’t bolt a proof beat onto the Ailing monk-authority script — the monk’s own authority is already the proof, and both proof kinds (living proof and the visiting doctor’s disbelief) lost to the proof-free control; keep Ailing scripts proof-free by default.
  TBF Exp 51 — Ailing proof beat · Aug 15, 2026
• avatar The incumbent hard-years monk script remains the Ailing Control; Health Is Fortune delivered in the monk frame but lost to it, so keep it as a secondary territory rather than the lead.
  TBF Exp 44 — Ailing monk frame territory · Aug 6, 2026
• avatar Sample new territories inside the avatar's proven frame before funding new speakers; the monk-frame retest separated territory from format where Exp 40 could not.
  TBF Exp 44 — Ailing monk frame territory · Aug 6, 2026
• avatar Write Ailing first-person scripts in the health-is-fortune territory: fortune is a morning without the weight; money luck arrives last.
  TBF Exp 40 — Ailing territory batch · Jul 30, 2026
• avatar Failed-remedies (crystals are passive, copper conducts) is the only runner-up worth a later isolation. Peer testimonial is parked.
  TBF Exp 40 — Ailing territory batch · Jul 30, 2026

The Cursed · TBF · 20 learnings

• avatar Lead the Cursed native primary text with the Blindsided lived-experience frame — it earned the read while the other seven provisional segments went unread on this surface.
  TBF Exp 82 — Provisional avatar discovery · Sep 2, 2026
• avatar Promote Blindsided to its own named segment with a dedicated follow-up test against the Cursed control before adopting the frame anywhere else.
  TBF Exp 82 — Provisional avatar discovery · Sep 2, 2026
• avatar Treat the seven unread segments as rejected on this surface rather than a ranked ladder of near-misses — only Blindsided carries a real verdict.
  TBF Exp 82 — Provisional avatar discovery · Sep 2, 2026
• avatar Open Cursed native-image primary text as a named person's origin story with a why-now beat — the named voice won and the anonymous household-confessional stakes voice lost, so do not lead with the anonymous stakes voice.
  TBF Exp 80 — Cursed origin voice · Sep 2, 2026
• avatar The winning why-now beat is pre-ownership — anchor the Cursed origin opening in having owned the object before, and skip affordability, betrayal, inheritance, and practitioner framings.
  TBF Exp 80 — Cursed origin voice · Sep 2, 2026
• avatar Match the envy-to-fortune explanation to The Cursed's own documented curse-and-block vocabulary instead of a neutral explanatory register — the avatar already names the problem in those words, and mirroring them lowered interpretation effort without changing the proposition.
  TBF Exp 77 — Cursed message isolation v2 1 replacement · Aug 29, 2026
• avatar Lead The Cursed's copper-and-protection proof with verification-before-belief — show the reader how to check the claim — rather than legend-led assertion that asks for belief on authority.
  TBF Exp 67 — Cursed message isolation v2 1 replacement · Aug 29, 2026
• avatar Inside the 17-traveler proof span, put the verify-first step before the legend so the reader's belief is earned, not assumed.
  TBF Exp 67 — Cursed message isolation v2 1 replacement · Aug 29, 2026
• avatar Carry the source billboard's full commercial stack into The Cursed dense-caption translations — concrete wearer and review counts, a stated price and savings, and an explicit guarantee — instead of trimming it to light proof and a vague availability nudge.
  TBF Exp 57 — Commercial credibility · Aug 22, 2026
• avatar For The Cursed, preserve concrete, observable pain and payoff language even when the product mechanism is spiritual — the concrete-bookends version is the arm the auction chose to run, so keep the specificity instead of reverting to abstract luck/drain phrasing.
  TBF Exp 56 — Concrete problem outcome · Aug 22, 2026
• avatar Open Cursed Champa scripts on the wealth-root / money-disappearing curse, not the classic original-curse thesis.
  TBF Exp 41 — Champa script territory · Jul 31, 2026
• avatar The Champa first-person message — envy framed as a curse that drains the root chakra — is a delivering creative for The Cursed; it carried the ad set on its own, so keep it as the production anchor.
  TBF Exp 37 — Cursed champa new control text message isolated · Jul 29, 2026
• avatar The Italian Cursed audience responds to the monk mechanism — the winning arm crossed the meaningful-delivery threshold in a greenfield geo with zero creative history. This unblocks Italian TBF scaling. Before shipping more Italian creatives, settle the translator decision here so every future Italian script uses the proven pipeline.
  TBF Exp 29 — Italian translation pipeline deepseek vs deepl · Jul 29, 2026
• avatar For The Cursed, the brochure block is the leak — replace it with teaching facts (copper-on-skin, the 500-monastery number) and close on “if you recognize this” so the monk never turns into a salesperson.
  TBF Exp 32 — Offer seam integration · Jul 25, 2026
• avatar Keep the Joan Ginther lineage character as the protagonist when porting the factory to The Cursed — the credentialed four-win figure carries, while freshly minted everyman winners (hairdresser, electrician, nurse’s aide, forklift operator, postal worker) never got traction.
  TBF Exp 30 — Cursed character factory translate · Jul 24, 2026
• avatar Brief future Cursed Deepstash clones into stacked_text_montage.
  TBF Exp 28 — Kinetic vs stacked layout · Jul 23, 2026
• avatar: Cursed The exact 'In one minute I can tell you…' diagnosis phrasing is confirmed hook doctrine for this avatar. Do not paraphrase it; the literal phrasing is the asset.
  TBF Exp 13 — One minute diagnosis hook refinement · Jul 11, 2026
• avatar What is unique to V2: it attributes the knowledge gap to the READER's personal, active ignorance ("you don't realize") rather than describing a mysterious external object ("the strange reason" / "the hidden pattern") or a weaker passive ("miss"). Personal active-ignorance framing beats both object-centric mystery and generic missed-knowledge.
  TBF Exp 5 — Mystical knowledge headline structure curses jetsetters · Jun 9, 2026
• avatar: Cursed Lead with a specific, curiosity-driven claim ("the strange reason women over 50 get unlucky") rather than a generic forbidden-secret opener ("most people think a curse looks like something dramatic") — Jetsetters respond to specificity and mystery, not vague secrecy
  TBF Exp 3 — Hook type and narrator voice curses jetsetters · May 19, 2026
• avatar: Cursed Forbidden Truth hooks were rejected across all scriptwriters for this audience — the "most spiritual teachers won't say this" / "most people think a curse looks dramatic" framing doesn't create enough specific curiosity for Jetsetters
  TBF Exp 3 — Hook type and narrator voice curses jetsetters · May 19, 2026

The Overlooked · TBF · 14 learnings

• avatar Keep the plane-crash dramatic-incident frame in The Overlooked's rotation — the Jeju-crash video still earns its place against today's landers and Sheet copy, so a near-miss disaster moment remains a live opener for this avatar.
  TBF Exp 72 — Jeju crash retest · Aug 30, 2026
• avatar When relaunching a proven 2025 dramatic-incident video, point it at the current lander pair and Sheet copy instead of its retired go-link — the frame survives the contract change.
  TBF Exp 72 — Jeju crash retest · Aug 30, 2026
• avatar The Overlooked's protection-luck scaffolding converts on the survival story itself — the death attribution is not the load-bearing element here.
  TBF Exp 58 — Walter named dead transfer · Aug 26, 2026
• avatar Keep the Overlooked Miranda message at full length — the compressed cut never got traction, so the compression cut something load-bearing; don’t trim this avatar’s proven message.
  TBF Exp 55 — Overlooked avatar miranda experiment message compression · Aug 21, 2026
• avatar For The Overlooked, don't bet the next round on more twist edits — the proven Control already owns that frame, so put the next challenger on a different frame or a different lever entirely.
  TBF Exp 48 — Overlooked first person contemporary twist variants · Aug 11, 2026
• avatar Do not delete the Overlooked valedictory close. Cutting it lost to relocating it.
  TBF Exp 36 — Overlooked miranda close structure · Aug 7, 2026
• avatar Keep the watch as the visible status cue on Overlooked factory males in this clothes-control lineage.
  TBF Exp 46 — Peter clothes control rerun · Aug 6, 2026
• avatar Keep the original pacing for The Overlooked testimonial footage — the 1.3x speed-up re-edit failed to deliver. Longer, more deliberate timing carries authenticity for this avatar. The tight-edit instinct (faster pacing equals better retention) is wrong here.
  TBF Exp 42 — Overlooked twist editing style · Aug 3, 2026
• avatar Write The Overlooked's seven-win streak as a solitary, private machine win (nickel-and-quarter slots) rather than a communal or charitable venue - the machine framing is what carried, and the audience's idea of 'lucky' here reads personal and mechanical, not social.
  TBF Exp 39 — Marge game of chance · Jul 30, 2026
• avatar Revisit the dignity constraint from Exp 002 - the modest, everyday machine-gambling story out-carried the respectable communal one, so don't assume a 'dignified' community win is the right frame for this avatar; a humble machine ritual may fit the audience's self-image better.
  TBF Exp 39 — Marge game of chance · Jul 30, 2026
• avatar Dolores is a viable second protection-luck character for the Overlooked roster. The character identity (67-year-old cafeteria worker, Dayton, church-van crash) connects — add Dolores to the character rotation.
  TBF Exp 38 — Dolores protection luck opening confirm · Jul 30, 2026
• avatar The 1-in-17-trillion, 1-in-18-septillion, and professor quote are not decorative — they are load-bearing credibility devices for the Overlooked avatar
  TBF Exp 7 — Septillion odds framing in Joan Ginther VSL · Jun 20, 2026
• avatar Removing the researcher/statistical proof block stopped delivery; the external proof asset appears necessary for the Overlooked avatar to accept the bracelet mechanism
  TBF Exp 6 — Scientific plausibility in Joan Ginther VSL · Jun 20, 2026
• avatar If proof is load-bearing, future scripts should keep a population-level or credentialed third-party finding rather than relying on the protagonist's private mystical explanation
  TBF Exp 6 — Scientific plausibility in Joan Ginther VSL · Jun 20, 2026

The Spiritually Curious Sceptic · TBF · 8 learnings

• avatar A visible presenter carries the Spiritually Curious Sceptic message better than message-led text; do not move SCS creative to slide formats by default.
  TBF Exp 59 — Scs format translation · Aug 26, 2026
• avatar Lead the Spiritually Curious Skeptic with the unexplained-setbacks wound; keep the 'someone close has shifted' UMP and the dated cases as the shared spine.
  TBF Exp 52 — Scs pain type · Aug 15, 2026
• avatar The one-minute diagnosis frame earns attention on the Spiritually Curious Skeptic — audiences watch it. But the frame earns attention without converting: the Skeptic stays curious but unconvinced.
  TBF Exp 23 — One minute diagnosis scs translate · Jul 19, 2026
• avatar The SCS conversion path needs the recognition reversal first — "It is not bad decisions, it is not bad karma" — before any diagnosis promise. Without the validation beat, the diagnosis feels like a sales pitch, not an expert consultation.
  TBF Exp 23 — One minute diagnosis scs translate · Jul 19, 2026
• avatar Keep SCS statics on the internal AdMaker benchmark pipeline. The vendor set did not match it.
  TBF Exp 19 — Stefanbrain vs admaker static bakeoff scs · Jul 17, 2026
• avatar The lost-discovery archive-ledger bridge is a confirmed credibility thread for the Spiritually Curious Skeptic. Future SCS scripts open with the rediscovered-archive frame (case ledgers, generations of observation, names change but the pattern doesn't).
  TBF Exp 16 — Scs lost discovery bridge · Jul 13, 2026
• avatar The SCS avatar now has a confirmed opener doctrine: Lost Discovery. The skeptical-scientist frame (14.5) is superseded, not invalidated — it remains an alternative bridge family.
  TBF Exp 16 — Scs lost discovery bridge · Jul 13, 2026
• avatar: Spiritually Curious Skeptic The Skeptical Scientist frame is confirmed for the skeptic avatar: credit the skeptic's worldview first ('she believed in neurons and probability'), then ask the one question that makes them go quiet ('when did it start?').
  TBF Exp 14.5 — Skeptic hook framing · Jul 8, 2026

Family · 145 learnings

• family Keep the Champa authority talking-head voiceover-only — the orchestral score bed and the score-plus-B-roll treatments never earned a read, so plain voiceover is the settled default for this format.
  TBF Exp 88 — Orchestral score · Sep 5, 2026
• family Do not add a motivational score or realistic cutaways to this monk authority format expecting a lift — the audience did not reward the extra production.
  TBF Exp 88 — Orchestral score · Sep 5, 2026
• family For the long Champa wealth-curse script, open with the symptom stack — name the repeated money setbacks and their emotional cost, then the envy diagnosis and the copper solution — instead of the direct wealth-curse promise; the recognition-first opening earned the run.
  TBF Exp 87 — Wealth curse symptom stack hook · Sep 5, 2026
• family Preserve the question-first curse reveal for the Champa curse-question anchor — opening with the direct curse question and the ordinary-jealousy-not-black-magic reframe beat the symptom-stack opener, so keep that grammar when remaking this anchor.
  TBF Exp 86 — Curse qa symptom stack hook · Sep 5, 2026
• family Restore the observable-payoff sentences in the close — a payoff the viewer can picture (“money feels easier to hold”) earns the read, while a payoff stated only as a date does not.
  TBF Exp 85 — Concrete outcome restoration · Sep 5, 2026
• family When you compress a source VSL, preserve the observable-outcome sentences and take the length savings from elsewhere — the Exp 14 compression cut load-bearing copy.
  TBF Exp 85 — Concrete outcome restoration · Sep 5, 2026
• family Keep the concrete opening/outcome bookends stacked on the commercial-credibility block when rebuilding the Cursed dense-caption script — the combined span stack beat the credibility block alone.
  TBF Exp 78 — Combined restoration · Sep 2, 2026
• family Winning spans from sibling isolations stack on a shared skeleton — promote the combined restoration as this family's new control and default to combining disjoint winners before testing narrower variations.
  TBF Exp 78 — Combined restoration · Sep 2, 2026
• family Make the June riptide, Cape May near-miss the new default protagonist for this news-doc family — a specific close-call water story holds The Overlooked where the Peter lightning win now falls flat.
  TBF Exp 71 — Peter news doc stories · Aug 30, 2026
• family Build the next news-doc script around a single vivid near-miss incident (one place, one moment) rather than a broad lightning/lottery win.
  TBF Exp 71 — Peter news doc stories · Aug 30, 2026
• family Apply a less-AI realism pass to the presenter's footage before shipping the next cut in this VSL family — a more natural-looking speaker holds the audience where an obviously-AI presenter loses it.
  TBF Exp 69 — Ginther speaker realism · Aug 30, 2026
• family: cursed-message-isolation Keep the swap surgical: change only the vocabulary span that diverges from the avatar's own wording, leaving the causal meaning and every surrounding byte intact — a faithful minimal edit outperformed the house register.
  TBF Exp 77 — Cursed message isolation v2 1 replacement · Aug 29, 2026
• family Keep the original ER Nurse pacing cut as the finished asset for this footage — a picture re-cut directed by an editing chief did not improve it, so leave the picture alone and stop chasing visual polish.
  TBF Exp 65 — Er nurse amoo editing chief · Aug 29, 2026
• family With the script and voice held byte-identical and only the picture changing, the re-cut changed the outcome not at all — the next lever on this VSL family is a message or voice change, not a visual edit.
  TBF Exp 65 — Er nurse amoo editing chief · Aug 29, 2026
• family Teach the two-pathway mechanism inside the reader's own experience — the second-person foil (“your turmeric half-worked, here's why”) beat every abstract frame, so make the mechanism a fact about the reader's own body, not a story about the joint.
  JA Exp 14 — FKC UMS vivid-presentation isolation on the Redmond skeleton · Aug 28, 2026
• family Drop stand-alone analogies — smoke-through-two-doors and the garden-gate hinge lost the reader, while concrete vehicles that reach into the reader's own cupboard (tour, dialogue, second-person foil) held.
  JA Exp 14 — FKC UMS vivid-presentation isolation on the Redmond skeleton · Aug 28, 2026
• family When the reader recognizes themselves in the mechanism (“my turmeric only half-worked”), the reversal lands — the reader-foil is the load-bearing move, not the analogy.
  JA Exp 14 — FKC UMS vivid-presentation isolation on the Redmond skeleton · Aug 28, 2026
• family Repairing the leaked writer-context passage in the Ageing r2-fo-06 documentary did not outdraw the bugged original; the leak is not what the auction was pricing.
  TBF Exp 60 — Ageing writer context repair · Aug 27, 2026
• family Keep full-frame visible Champa (authority_talking_head) as the default carrier for the Sceptic unexplained-setbacks message; the corner-authority and kinetic-text packages of the identical script were not funded.
  TBF Exp 59 — Scs format translation · Aug 26, 2026
• family The named-dead beat did not transfer from Dolores to Walter — hold it as a Dolores-specific proof beat rather than promoting it to the protection-luck family brief.
  TBF Exp 58 — Walter named dead transfer · Aug 26, 2026
• family In the dense-caption family the closing commercial block is the load-bearing sell: a proof/value/offer/guarantee close beat a lighter proof-and-availability close on the same creative, so do not soften the sell when localizing a source billboard.
  TBF Exp 57 — Commercial credibility · Aug 22, 2026
• family In the dense-caption family, SwipeSmith translations must keep the source's concrete, observable losses and improvements at the opening and outcome bookends — translating them into abstract belief language weakened the creative, and concreteness is a keeper, not a failed diagnosis.
  TBF Exp 56 — Concrete problem outcome · Aug 22, 2026
• family Render monk-authority diagnosis as the corner PIP over the animated educational bed, not the plain full-frame talking head.
  TBF Exp 54 — Corner vs authority cadence repair · Aug 20, 2026
• family Keep the educational bed under the corner PIP — the full-frame talking head does not carry the diagnosis.
  TBF Exp 54 — Corner vs authority cadence repair · Aug 20, 2026
• family Don't treat the exact original winner bytes as untouchable — a refreshed finish (re-edit plus a new voice read) on the proven Marge testimonial out-earned the untouched original, so refresh the finish of a proven testimonial rather than locking its original cut as sacred.
  TBF Exp 53 — Marge amoo edit new voice · Aug 18, 2026
• family This win bundled the edit and the voice together, so it proves a finish refresh works but not which lever carried it — split editing and voice into separate variants next time before promoting either component as a rule.
  TBF Exp 53 — Marge amoo edit new voice · Aug 18, 2026
• family Do not graft an isolation-round span winner verbatim into a changed-register skeleton — the two-pathway-king span that won at the literate register lost to the heritage 'restore the forgotten king' line at the plain register; compose-and-verify before promoting any span winner into a brief.
  JA Exp 13 — FKC champion assembly via UMS graft · Aug 15, 2026
• family An isolation-round span verdict is register-specific — re-verify the span inside the new skeleton before writing it into the brief rather than composing winners blindly.
  JA Exp 13 — FKC champion assembly via UMS graft · Aug 15, 2026
• family Name the SCS wound as a pattern of setbacks that will not explain themselves — that felt pain beat fortune-turned and unmatched-potential.
  TBF Exp 52 — Scs pain type · Aug 15, 2026
• family In this family the monk-authority frame is self-proving — proof is a burden, not an asset, so don’t import the proof-heavy structure from other avatars’ winners into the monk frame.
  TBF Exp 51 — Ailing proof beat · Aug 15, 2026
• family Keep a named person who did not make it in the first near-death of protection-luck scripts — the survival only reads as real when someone else died in their place.
  TBF Exp 50 — Dolores carol dead substitute · Aug 15, 2026
• family Stop writing the solo-survival version of the church-van wreck; the named-dead substitute is the load-bearing proof beat, not decorative color.
  TBF Exp 50 — Dolores carol dead substitute · Aug 15, 2026
• family Keep the existing first-person-contemporary-twist Control as the finishing standard — three fresh editor variants (V1/V2/V3) all failed to earn traction against it, so this family's Control is already at its ceiling.
  TBF Exp 48 — Overlooked first person contemporary twist variants · Aug 11, 2026
• family Keep 'write it as history' as the FKC doc-VSL brief default; the contemporary-subject twin in the same omniscient voice ran level with the historical Control and gave no reason to drop the history.
  JA Exp 12 — FKC omniscient voice vs historical subject (the confound in Exp 007) · Aug 8, 2026
• family Keep Miranda's five-sentence valedictory and place it immediately before the scarcity block. That close structure won.
  TBF Exp 36 — Overlooked miranda close structure · Aug 7, 2026
• family Do not leave the valedictory at the original end position. The scarcity-seam placement is the winning close.
  TBF Exp 36 — Overlooked miranda close structure · Aug 7, 2026
• family Do not revert Peter to the unchanged-clothes Control. The former V4 treatment beat baseline.
  TBF Exp 46 — Peter clothes control rerun · Aug 6, 2026
• family Family-healing and failed-remedies were not funded in the monk frame either, after starving as first-person scripts in Exp 40; retire both territories for the Ailing.
  TBF Exp 44 — Ailing monk frame territory · Aug 6, 2026
• family Write FKC doc-VSLs in plain language around FK 4, not the literate FK 7-8 register the Control inherited.
  JA Exp 11 — FKC doc-VSL reading-level isolation · Aug 3, 2026
• family Give every future JA long-form brief an explicit target reading grade. Default it to the winning plain-language register.
  JA Exp 11 — FKC doc-VSL reading-level isolation · Aug 3, 2026
• family Keep the proven back half byte-for-byte from the pivot line onward; in this family the opener is the only part that carries a new idea.
  JA Exp 10 — FKC turmeric-verdict on the scale surface (construction batch) · Aug 3, 2026
• family Apply the propped-phone treatment (6% overscan, ±12px two-axis sway, ±0.5° rotation, mild sensor grain, ±1.5% exposure drift) as the default render finish for AI talking-head testimonials in this family. Moderate visual imperfection at this dose outperforms clean renders.
  TBF Exp 43 — Marge visual imperfection · Aug 3, 2026
• family Do not use subliminal-only imperfection — the barely-conscious wobble was Meta-starved. The treatment must be consciously visible to work as an authenticity signal; sub-perceptual wobble earns no credit.
  TBF Exp 43 — Marge visual imperfection · Aug 3, 2026
• family Do not push past propped-phone into active-handheld or glitch territory — both were rejected harder than the clean Control. Visual imperfection has a sharp dose-response curve; the winning band is narrow.
  TBF Exp 43 — Marge visual imperfection · Aug 3, 2026
• family The visual channel now joins the audio channel as a proven authenticity lever — Exp 034 confirmed spoken disfluencies; Exp 043 confirms visual imperfection. Every future AI testimonial render should carry both disfluencies AND the propped-phone treatment. The two-channel finding is now closed.
  TBF Exp 43 — Marge visual imperfection · Aug 3, 2026
• family The editing lever is real — a decisive delivery split on identical script and character — but the winning direction is original pacing, not tighter cuts. For this testimonial family, a clean tripod-stable video at natural speed is the delivery leader.
  TBF Exp 42 — Overlooked twist editing style · Aug 3, 2026
• family Explain Joint Ascent through the two-pathway king — a checkable hero-ingredient mechanism the reader can point at — rather than the heritage 'restore the forgotten king' line.
  JA Exp 9 — FKC UMS isolation test: which Unique Mechanism of the Solution explains Joint Ascent best · Aug 2, 2026
• family Do not keep the four-jobs, turmeric-completion, or one-capsule-replaces-the-shelf explanations as the default solution story — the two-pathway king is the frame that carried the read.
  JA Exp 9 — FKC UMS isolation test: which Unique Mechanism of the Solution explains Joint Ascent best · Aug 2, 2026
• family Park luck-accumulates, relational-source, hard-work-behind, one-minute diagnosis, childhood-words, and smile-wrapped-words as Champa script territories.
  TBF Exp 41 — Champa script territory · Jul 31, 2026
• family Original Curse Thesis is the runner-up only. Do not keep it as the default franchise spine.
  TBF Exp 41 — Champa script territory · Jul 31, 2026
• family The back-half selling question is still open — the direct-offer and proof closes were never funded, so do not read this run as the elegiac close beating them.
  JA Exp 8 — Doc-VSL back-half strategy test (ROAS iteration on 007's winner) · Jul 30, 2026
• family Keep the elegiac zero-sell close as the working back half for now — it held the audience — but treat the offer/proof question as untested rather than answered.
  JA Exp 8 — Doc-VSL back-half strategy test (ROAS iteration on 007's winner) · Jul 30, 2026
• family When the front half is byte-identical, give the challenger back halves a structurally different shape (a hard offer/CTA block, not a near-sibling retelling) or the audience cannot tell them apart and the test stalls.
  JA Exp 8 — Doc-VSL back-half strategy test (ROAS iteration on 007's winner) · Jul 30, 2026
• family Do not default new Ailing characters to peer testimonial, caregiver drain, or household-blessing. Those territories lost.
  TBF Exp 40 — Ailing territory batch · Jul 30, 2026
• family Treat the game of chance as a load-bearing copy decision in the ginther-vsl family, not window dressing - swapping bingo for slots flipped the result, so choose the game deliberately and let it carry part of the story's meaning instead of defaulting to a safe communal setting.
  TBF Exp 39 — Marge game of chance · Jul 30, 2026
• family The survival-first opening is NOT a universal protection-luck rule. It worked for Walter McCathie (Exp 021) but lost to the win-first opening on Dolores. Do not hard-code a single opening shape into the factory brief — every new protection-luck protagonist needs their own opening-beat pair.
  TBF Exp 38 — Dolores protection luck opening confirm · Jul 30, 2026
• family When creating new protection-luck characters, always test opening-beat order with a fresh pair. Never assume the prior character winner transfers — the Exp 021 Walter result stands but is Walter-scoped.
  TBF Exp 38 — Dolores protection luck opening confirm · Jul 30, 2026
• family Use DeepL, not DeepSeek V4 Pro, to translate proven English TBF scripts into Italian. DeepL's Italian output got 3x the delivery volume of the same script through DeepSeek — raw pipeline, no human post-edit on either arm.
  TBF Exp 29 — Italian translation pipeline deepseek vs deepl · Jul 29, 2026
• family Don't assume the LLM that writes great English TBF scripts also produces the best Italian translations. DeepSeek V4 Pro is the proven English scriptwriter but its raw Italian output got starved by Meta. Translation fidelity to the persuasive structure matters more than LLM fluency.
  TBF Exp 29 — Italian translation pipeline deepseek vs deepl · Jul 29, 2026
• family Prefer the short one-liner cadence over compact 11-paragraph blocks for this avatar. The rhythm is load-bearing.
  TBAB Exp 5 — Primary text entry thesis · Jul 25, 2026
• family Do not lead with ancient-sacred-frequency exposition, a damaging admission, or a question-first open. Those lost to Stanford-in-one-liners.
  TBAB Exp 5 — Primary text entry thesis · Jul 25, 2026
• family An unmotivated, claim-free spectacle shot in the first ~3 seconds of a Cursed music-video asset earned its render cost decisively — the platform directed nearly all delivery to the spectacle arm while ignoring the lyric-motivated Control. The visual-impossibility signal (production capability the prospect cannot get elsewhere) is itself the hook.
  TBF Exp 33 — Unmotivated spectacle opening · Jul 25, 2026
• family When producing a Cursed realistic music video, brief one pure-spectacle opening shot (abstract, visually impossible, zero claims) alongside lyric-driven opens. The spectacle does not need narrative justification to work.
  TBF Exp 33 — Unmotivated spectacle opening · Jul 25, 2026
• family Keep the monk message in teaching voice through the close — don’t switch into “we made this bracelet / it’s named / it brings blessings” shop voice; the audience converts while still being taught, not after being sold.
  TBF Exp 32 — Offer seam integration · Jul 25, 2026
• family Treat the character factory as character-specific, not skeleton-specific: the 14-beat skeleton is portable, but its power to mint NEW characters does not transfer across avatars, so reuse the avatar’s proven protagonist rather than regenerating one.
  TBF Exp 30 — Cursed character factory translate · Jul 24, 2026
• family Default Deepstash-class Cursed music-only ads to the IG stacked layout — full script always visible under a top media band.
  TBF Exp 28 — Kinetic vs stacked layout · Jul 23, 2026
• family Do not keep progressive full-bleed kinetic captions as the default for this class. Stacked won.
  TBF Exp 28 — Kinetic vs stacked layout · Jul 23, 2026
• family Write Cursed song lyrics from specific spoken soft-curse phrases the audience recognizes ('must be nice', 'you're so lucky', said with salt) rather than conceptual thesis statements. The Cursed audience responds to the exact words they've heard, not abstract ideas about shadow or jealousy.
  TBF Exp 27 — Cursed song thesis · Jul 23, 2026
• family A proven winning song thesis (Where the Shadow Ends, previously a strong standalone winner) can get starved when placed next to a stronger song thesis in the same ad set. Do not treat a past winner as a permanent Control — every new song batch is a real head-to-head.
  TBF Exp 27 — Cursed song thesis · Jul 23, 2026
• family The song thesis matters more than format familiarity for Cursed music videos. When briefing new Cursed songs, start from a real spoken phrase the prospect has heard, not from a conceptual frame like 'internal shadow' or 'jealous eye.'
  TBF Exp 27 — Cursed song thesis · Jul 23, 2026
• family Keep specific dated cases with ledger detail in every future SCS lost-discovery script. Named people, years, and the archive-ledger texture are the load-bearing proof.
  TBF Exp 26 — Scs lost discovery proof type · Jul 21, 2026
• family Do not replace those named cases with a generational pattern-of-many. The cases are the asset, not scaffolding to cut.
  TBF Exp 26 — Scs lost discovery proof type · Jul 21, 2026
• family Do not close the lost-discovery bridge on a final-entry identification pivot. That closer lost to named-case proof.
  TBF Exp 26 — Scs lost discovery proof type · Jul 21, 2026
• family Next move inside this frame is the threat TARGET, not a new phrase type. Phrase type is settled: envy.
  TBF Exp 25 — Social threat phrase type · Jul 21, 2026
• family Do not swap in 'why can't you be more like your sister' or 'that spiritual stuff won't pay the bills' as the default named phrases.
  TBF Exp 25 — Social threat phrase type · Jul 21, 2026
• family The Cursed hears "your luck has been stolen" and converts; the Skeptic hears "someone close to you has shifted" and watches but does not buy. The diagnosis frame is bound to the Cursed theft thesis for conversion — it does not transfer.
  TBF Exp 23 — One minute diagnosis scs translate · Jul 19, 2026
• family Do not switch SCS static supply to StefanBrain. The bake-off chose the internal concepts.
  TBF Exp 19 — Stefanbrain vs admaker static bakeoff scs · Jul 17, 2026
• family The Joan Ginther script skeleton is a confirmed character factory: it produced a winning protagonist independent of Joan and across a different luck domain. The factory works by swapping the protagonist inside the fixed beat sequence, not by cloning one character type.
  TBF Exp 17 — Ginther character batch · Jul 14, 2026
• family Four starved characters are four auction rejections, not untested rivals — Earl, Rosa, Hazel, and Control Joan herself were each tried and declined.
  TBF Exp 17 — Ginther character batch · Jul 14, 2026
• family Five starved challengers are five rejected rival framings. The incumbent did not merely survive — Meta tried the alternatives and rejected them.
  TBF Exp 13 — One minute diagnosis hook refinement · Jul 11, 2026
• family Declan is the confirmed narrator voice for this realistic documentary family. The earlier Andrew Zhang call was the first-week leader, not the matured winner.
  LLM Exp 3 — Realistic 3x4 narrator voice · Jul 10, 2026
• family A voice that leads in the first week can still lose once the auction settles. Keep the matured winner as the production default.
  LLM Exp 3 — Realistic 3x4 narrator voice · Jul 10, 2026
• family Cheaper Pixar render paths (Kling KF, Seedance Pro KF) could not prove their visual quality holds up well enough against the happy-horse default — and the test has only run 3 days. Don't abandon the cost-decomposition thesis yet.
  TBAB Exp 4 — Pixar scene renderer cost decomposition · Jul 10, 2026
• family 450 words is the confirmed compression target for the Richard/Champa message family. The 767-word depth adds no delivery value; the 250-word cut loses it.
  TBF Exp 14 — Richard champa message compression · Jul 10, 2026
• family Two starved arms bracket the winner: too long (767w) rejected, too short (250w) rejected. The auction mapped the viable length band in one experiment.
  TBF Exp 14 — Richard champa message compression · Jul 10, 2026
• family Delivery without conversion (V2 Pattern Evidence was served but produced no revenue) is a second, independent rejection layer — the platform delivered it and the market declined it. Both layers point at the same winner.
  TBF Exp 14.5 — Skeptic hook framing · Jul 8, 2026
• family Jealous Eye (jealousy as the attaching force) is the confirmed villain-source decomposition for hook 5. Future hook-5 scripts open on the jealous-person frame.
  TBF Exp 11 — Hook 5 villain source decomposition · Jul 7, 2026
• family Four starved arms are four auction rejections. The delivered winner already settled the contest; no fresh pair needed.
  TBF Exp 11 — Hook 5 villain source decomposition · Jul 7, 2026
• family Kira-collab refinement produced the only variant Meta gave real delivery to. For the Hidden Sound Map concept, the Kira-refined version (86.30) dramatically outperformed the direct-written version (18.28, NO-DELIVERY). This suggests Kira sharpened the emotional logic, pacing, or CTA handoff in ways Meta's algorithm rewarded.
  TBAB Exp 3 — Spiritual seeker pixar soulgem kira vs direct · Jul 6, 2026
• family Concept matters. The Hidden Sound Map premise (knowledge-discovery arc) attracted more total delivery than The Child Who Remembered the Silence (human-story arc). Kira may have amplified a concept that already had appeal rather than rescuing a weak one.
  TBAB Exp 3 — Spiritual seeker pixar soulgem kira vs direct · Jul 6, 2026
• family Don't make Kira-collab the default for all TBAB Pixar SoulGem scripts yet. Pay the Kira cost when the concept is a knowledge-discovery arc; test whether Kira adds value for emotional human-story arcs separately before extending the default.
  TBAB Exp 3 — Spiritual seeker pixar soulgem kira vs direct · Jul 6, 2026
• family 1.3x playback speed is the confirmed delivery pacing for the Joan Ginther VSL family. Future Ginther renders ship at 1.3x by default.
  TBF Exp 9 — Joan Ginther VSL playback speed test · Jul 2, 2026
• family Control won early; the astronomical/statistical odds framing appears to be the active proof asset
  TBF Exp 7 — Septillion odds framing in Joan Ginther VSL · Jun 20, 2026
• family Removing all statistical impossibility framing eliminated delivery entirely
  TBF Exp 7 — Septillion odds framing in Joan Ginther VSL · Jun 20, 2026
• family Wait for mature 90-day data, but do not cut the odds framing from future scripts until a clean test disproves it
  TBF Exp 7 — Septillion odds framing in Joan Ginther VSL · Jun 20, 2026
• family An under-$50 spend on Var A means Meta did not find an audience for mystical-only attribution — the platform signal rejects removing proof
  TBF Exp 6 — Scientific plausibility in Joan Ginther VSL · Jun 20, 2026
• family Wait for mature 90-day data before finalizing, but early signal favors keeping the Daniel Reyes / researcher proof scene
  TBF Exp 6 — Scientific plausibility in Joan Ginther VSL · Jun 20, 2026
• family For this Jetsetters/Curses Ted Talk body, the current signal favors opening with a fast authority diagnosis: "In one minute I can tell you if your luck has been stolen." The viewer gets a clear promise and a credible diagnostic frame before the mechanism begins.
  TBF Exp 12 — Ted talk hook framing cursed jetsetters · Jun 17, 2026
• family The symptom-triplet hook had strong historical spend in the lagged 90-day table, so do not throw it away; treat this as a recency correction, not proof that symptom-triplets never work.
  TBF Exp 12 — Ted talk hook framing cursed jetsetters · Jun 17, 2026
• family Keep the next test narrow: authority diagnosis vs symptom-triplet, same monk body, same audience, fresh ad set.
  TBF Exp 12 — Ted talk hook framing cursed jetsetters · Jun 17, 2026
• family The monk-secret opener is actively hurting performance — not just neutral, worse than nothing. Drop the opener entirely. The mechanism line IS the hook.
  JA Exp 5 — Monk-secret opener test · Jun 14, 2026
• family JA's winning template is now 2 lines: mechanism + closer. Don't add an intrigue preamble — "Knotted energy flow causes stiffness - not age" is strange and interesting enough to stop the scroll on its own.
  JA Exp 5 — Monk-secret opener test · Jun 14, 2026
• family If you must use an opener for a different format, doctor/authority framing (V2) was the best of the losing options — but it still lost to no opener at all.
  JA Exp 5 — Monk-secret opener test · Jun 14, 2026
• family The "secret" framing specifically may create distance rather than connection. Information asymmetry with institutional authority (doctors, healers) worked better than secret-keeping by monks.
  JA Exp 5 — Monk-secret opener test · Jun 14, 2026
• family A sibling mechanism beat the Control → Energy Flow is not uniquely the asset. The win belongs to the "mystical felt cause" category, not that specific phrase.
  JA Exp 4 — Mystical mechanism family test · Jun 12, 2026
• family "Knotted" may outperform "Blocked" because it implies something that can be untied — more actionable and visceral than "blocked."
  JA Exp 4 — Mystical mechanism family test · Jun 12, 2026
• family Trapped Tension and Trapped Life Force underperformed — "trapped" framing may be too static/passive compared to "knotted" or "frozen" which imply a reversible state.
  JA Exp 4 — Mystical mechanism family test · Jun 12, 2026
• family Open with a biological reframe before introducing the authority figure — tell them what stress actually is in the body before Champa arrives, so the monk's framework feels like an answer rather than a pitch
  KB Exp 4 — Stress hook type test · Jun 11, 2026
• family Repositioning stress as a broken biological signal rather than a life circumstance gives the mechanism more authority — it makes the Shen Men point feel like a real physiological explanation rather than a wellness concept
  KB Exp 4 — Stress hook type test · Jun 11, 2026
• family Note: this ad set is not yet profitable (0.81x overall ROAS) — the hook winner may still shift as Meta continues testing
  KB Exp 4 — Stress hook type test · Jun 11, 2026
• family Open the artisan-monk frame by naming the reader's problem as theft by envy, not by teaching abstract curse mechanics. "Fortune was stolen by someone else's envy" gives the cursed reader a concrete villain and a concrete loss.
  TBF Exp 10 — Artisan monk pre hooks curses jetsetters · Jun 9, 2026
• family Make the product feel purpose-built in the first line. Hook 5 works because the bracelet is introduced as made specifically for that stolen-fortune problem before the shared monk explanation begins.
  TBF Exp 10 — Artisan monk pre hooks curses jetsetters · Jun 9, 2026
• family Pair monk craft visuals with one simple object in frame. The winning opening keeps attention on the bracelet being worked on, instead of splitting attention across monastery atmosphere or general ritual explanation.
  TBF Exp 10 — Artisan monk pre hooks curses jetsetters · Jun 9, 2026
• family The asset is NOT the curiosity structure generically, and NOT the "strange reason" phrase specifically — Control (strange reason) was beaten into runner-up only in early logging; 90-day data puts Control at $7.92, i.e. noise.
  TBF Exp 5 — Mystical knowledge headline structure curses jetsetters · Jun 9, 2026
• family The asset is NOT "missed-knowledge" as a category — V3 ("hidden pattern") and V5 ("the one thing... miss"), same family, were equal last.
  TBF Exp 5 — Mystical knowledge headline structure curses jetsetters · Jun 9, 2026
• family Implication for next batch: lock the "don't realize" personal-ignorance frame as the new control and test refinements of it (e.g. "don't realize" vs "have no idea" vs "don't know") rather than substituting alternative curiosity structures.
  TBF Exp 5 — Mystical knowledge headline structure curses jetsetters · Jun 9, 2026
• family The "not age" anti-fatalism reversal is doing real work — both winners contain it; all three variants without it failed.
  JA Exp 3 — Static ad age-fatalism reversal test · Jun 6, 2026
• family Energy Flow + "not age" beats Inflammation + "not age" — the mystical/felt cause outperforms the clinical cause when paired with the age-fatalism reversal.
  JA Exp 3 — Static ad age-fatalism reversal test · Jun 6, 2026
• family Energy Flow alone (without "not age") is not enough — the mechanism needs the belief reversal to land.
  JA Exp 3 — Static ad age-fatalism reversal test · Jun 6, 2026
• family Anti-age alone (without a specific alternative cause) is not enough — readers need something to replace age with, not just a denial.
  JA Exp 3 — Static ad age-fatalism reversal test · Jun 6, 2026
• family Generic symptom promise without mechanism or reversal does not perform — the ad needs either a named cause or the age-fatalism reversal to win.
  JA Exp 3 — Static ad age-fatalism reversal test · Jun 6, 2026
• family First person monk voice consistently outperformed 3rd person narration — the same body script (Richard vs Richard Hook 2) only scaled when it had a 1st person monk hook bolted on
  TBF Exp 3 — Hook type and narrator voice curses jetsetters · May 19, 2026
• family Shorter scripts (410 words) can outperform longer ones (767 words) when the hook does the heavy lifting — Mark Hook 2 was the shortest script and the top performer
  TBF Exp 3 — Hook type and narrator voice curses jetsetters · May 19, 2026
• family Authority UGC style (speaking directly to camera as a practitioner) outperformed First Person Contemporary VSL for the voodoo priestess character — the format matches the authority claim
  TBF Exp 4 — Authority style and hook voodoo priestess curses jetsetters · May 12, 2026
• family The winning script opens with a concrete, observable claim ("it comes for one thing at a time — first the money, then the people, then the health") rather than an abstract question or philosophical statement
  TBF Exp 4 — Authority style and hook voodoo priestess curses jetsetters · May 12, 2026
• family Using tradition-specific terminology (awanga, not just "curse") strengthened the authority — it signals real expertise rather than generic spirituality
  TBF Exp 4 — Authority style and hook voodoo priestess curses jetsetters · May 12, 2026
• family Urgency framing tied to the curse mechanic ("every year you wait is a year they win" / "what takes three weeks to remove in the first year takes three months in the fifth") is more compelling than generic scarcity ("only 300 bracelets left")
  TBF Exp 4 — Authority style and hook voodoo priestess curses jetsetters · May 12, 2026
• family Even within the same Authority UGC style + Direct Pain Point hook, the Q&A frame ("are curses real?") lost to the assertion frame ("people think they know what a curse looks like") — starting by challenging the audience's assumption created more engagement than answering a question nobody asked
  TBF Exp 4 — Authority style and hook voodoo priestess curses jetsetters · May 12, 2026
• family A named doctor endorsement and a personal discovery story both keep people reading but don't close the sale — engagement and conversion are different things
  TBAB Exp 2 — Post hook framing test · Apr 29, 2026
• family Every other hook approach (asking a question, leading with science, leading with emotion) couldn't even get attention in the first place against secrecy
  TBAB Exp 1 — Hook framing test · Apr 24, 2026
• family Emotional mechanism framing ("blocked energy flow") can beat authority framing ("King of Medicine") when the mechanism maps directly to the felt experience of the target avatar.
  JA Exp 2 — Energy Flow big idea (current winner)
• family "Ritual" > "formula" / "supplement" in ad copy — implies transformation, not just treatment.
  JA Exp 2 — Energy Flow big idea (current winner)
• family The strongest ad framing tied Tibetan tradition to a named authority ingredient: Terminalia chebula as the "King of Medicine."
  JA Exp 1 — big idea angle test
• family Specific ingredient authority beat broader origin frames, ancient/modern inflammation framing, monk-secret framing, and the high-altitude joint-problem angle.
  JA Exp 1 — big idea angle test
• family Parasympathetic reawakening + ancient ritual + demographic callout (women over 40) outperformed all EMF/physics-based framings
  KB Exp 3 — Primary text mechanism test
• family EMF disruption as the core mechanism underperformed across all three symptom angles — the audience may not buy the electromagnetic interference narrative
  KB Exp 3 — Primary text mechanism test
• family Demographic specificity ("women over 40") may be a significant lever — only the winner used it, so it's confounded with mechanism choice
  KB Exp 3 — Primary text mechanism test
• family Physics-heavy language (Lenz's Law, micro-currents, neodymium magnets) did not improve performance — jargon may create distance rather than credibility
  KB Exp 3 — Primary text mechanism test
• family Neck pain is the best-performing tension-location hook — the pillow/stress misattribution is familiar, low-friction, and strong
  KB Exp 2 — Tension location test
• family Chest tightness is a strong #2 — the heart/stress misattribution creates strong recognition
  KB Exp 2 — Tension location test
• family Shoulder tension validates the framework for desk workers — common + correctly-attributed symptoms still perform well
  KB Exp 2 — Tension location test
• family Stomach knots outperformed hands, jaw, and head — gut-stress connection resonates with this demographic
  KB Exp 2 — Tension location test
• family Clenched fists underperformed despite being the most "checkable right now" — instant visibility doesn't guarantee instant recognition
  KB Exp 2 — Tension location test
• family Jaw pain is one of the weakest hooks — the dental/TMJ misattribution doesn't create enough recognition
  KB Exp 2 — Tension location test
• family Head pressure is weakest — the screen/bluelight misattribution is too expected
  KB Exp 2 — Tension location test

Ad · 14 learnings

• ad For Walter, keep the survival-first opening — his solo scaffolding escape already carries the proof; the named-dead substitute does not earn its place in his story.
  TBF Exp 58 — Walter named dead transfer · Aug 26, 2026
• ad Dress Peter McCathie with the Omega watch on the current clothes. That presentation won.
  TBF Exp 46 — Peter clothes control rerun · Aug 6, 2026
• ad For Dolores: lead with the ticket win and the forward-reference to "three times I should have died," then rewind through the escapes. The win-first hook carried both arms and won the head-to-head.
  TBF Exp 38 — Dolores protection luck opening confirm · Jul 30, 2026
• ad Read this as a single-creative ad set, not a six-way test: the arms were the same message with only punctuation-level differences, so 'V4 won' carries no phrasing signal to copy — the message itself is the asset, not any one arm.
  TBF Exp 37 — Cursed champa new control text message isolated · Jul 29, 2026
• ad For protection-luck characters, open on the near-death event, not the win. Walter's scaffolding collapse is the scroll-stopper; the ticket is the payoff. Leading with the win first — the Exp 017 rule — is specific to money-luck characters and loses audience attention when the luck domain is survival.
  TBF Exp 21 — Walter opening beat · Jul 19, 2026
• ad The survival-first structure earns dramatically more audience attention than win-first in a clean head-to-head with identical production and the same character identity. The audience stays because the near-death event is inherently more gripping than a lottery announcement.
  TBF Exp 21 — Walter opening beat · Jul 19, 2026
• ad When the character factory produces a new protection-luck protagonist, hard-code the opening shape: near-death event first, ticket payoff second. Do not re-derive this per script — it is now a locked structural brief for the Overlooked avatar.
  TBF Exp 21 — Walter opening beat · Jul 19, 2026
• ad INT-03 is the concept to clone first when this bake-off is reused.
  TBF Exp 19 — Stefanbrain vs admaker static bakeoff scs · Jul 17, 2026
• ad Marge (diner waitress with seven bingo jackpots in 14 months, peer-proximity of the luck venue) is a confirmed formula character — the one protagonist the auction scaled from this six-way batch.
  TBF Exp 17 — Ginther character batch · Jul 14, 2026
• ad 540p downscaled video could not attract meaningful delivery against the same file at 720p in the McCathie pair, but the test is only 3 days old and the result covers only one of three planned replication pairs.
  TBF Exp 15 — Vsl delivery resolution 540p · Jul 10, 2026
• ad Cannot interpret a voice/framing test without delivery; the ad set is active but has not yet accumulated measurable spend
  TBF Exp 8 — Modest believable peer in Joan Ginther VSL · Jun 20, 2026
• ad Do not rewrite Joan's modest peer framing until delivery produces a signal
  TBF Exp 8 — Modest believable peer in Joan Ginther VSL · Jun 20, 2026
• ad If delivery remains absent, treat the variant package as a Meta rejection and return to the proven control voice
  TBF Exp 8 — Modest believable peer in Joan Ginther VSL · Jun 20, 2026
• ad Once spend arrives, the priority comparison is Control vs Var A (isolated voice change) before evaluating Var B (full rewrite)
  TBF Exp 8 — Modest believable peer in Joan Ginther VSL · Jun 20, 2026