Cocoa + Nattokinase launch · research
Cocoa + Nattokinase: Angles & Launch Landscape
v1, Oct 27 2026. The candidate angles scored against 4,796 coded ads, set against the filed calendar, with the compliance firewall. Numbers and method are in the research dossier; the creatives in the swipe gallery; the sends on the launch hub. No campaign copy.
1 · Product at a glance
| Component | Per serving | Status |
|---|---|---|
| Cocoa base | 1,200 mg cocoa flavanols, 600 mg epicatechin (label of the current cocoa) | label figures |
| Nattokinase | 10,800 FU | both options |
| L-citrulline | in; dose not set | Option B confirmed Oct 27 |
| Preparation | warm or cold, never hot | heat stability primary: about 50 °C; bench test pending |
For the cocoa buyer who already reaches for nattokinase next (13–16% of cocoa buyers' next purchases), and for the nattokinase searcher (“nattokinase” 251,619 a month) who has never been given a reason to drink it. No label, no name, no price yet.
2 · How the market sells it: vs. our play
Three ingredients, three separate markets, no shared voice. Nattokinase (1,515 ads) is sold on clots and plaque (48.1%) and men's performance (21.3%), lanes a supplement may not claim, mostly from persona pages and one ED advertorial operator. Cocoa (3,163 ads) is sold on the morning ritual, energy and doctor-voice video, and its biggest script (stem cells) is saturating. Citrulline (98 longest-running ads; 11,400 in the index) is sold to the gym. Toku (20 live ads) is the only one in a cup, and it sells nattokinase in chocolate flavour on the dose gap.
The open ground: the story that joins them. 0 ads mention watermelon; 14 mention the Nobel Prize, all of them cocoa; 6 cocoa ads mention nattokinase; 0 citrulline ads mention cocoa.
3 · The candidate angles
| Angle · where it lives | What the data says | Gate | Use | |
|---|---|---|---|---|
| A | Japan's breakfast, in the cup you already drink the identity, from the reveal (#4) on | Natto's thousand-year story is told in 965 of 1,515 nattokinase ads (3.8% lead with it); the morning cup is in 1,106 of 3,163 cocoa ads. Nobody joins them: 6 cocoa ads mention nattokinase. Toku is in a sachet with chocolate flavouring and no flavanol dose. | Never 'synergy', never 'better together'; each ingredient keeps its own job. The heat line waits for the bench test. | Lead identity for #4 to #8; the product is named once the label exists. |
| B | The two places your checkup never looks the frame for the arc (#1 sent) | The field sells the numbers (0.6% of nattokinase ads, 3.3% of cocoa ads lead with cholesterol, BP or the checkup) or the fear (48.1% clots and plaque). The vessel wall is named in 281 cocoa ads and 543 nattokinase ads. | Never 'clears plaque' or 'dissolves clots'; the wall and what forms inside it, never a disease outcome. | The frame every send hangs on; slot 01 opened it, slot 02 spends the wall. |
| C | The molecule they sold you instead slot 02 (citrulline) | Citrulline is sold to the gym (23.5% pump) and on generic blood flow (52.0%). 0 of 4,796 ads mention watermelon; 0 citrulline ads mention the Nobel Prize; 2 mention a checkup. The arginine argument (gut breakdown, VINTAGE MI) is unclaimed. | VINTAGE MI is heart-attack patients only; never 'arginine harms healthy people', never a brand. No citrulline dose of ours until the label. | Slot 02's secret; later the reason citrulline is in the cup at all. |
| D | The dose nobody tested slot 03 (e-book) | Contested: 393 nattokinase ads print 10,800 FU and Toku's live static says 'Most Nattokinase Is Underdosed… the dose used in a 12-month study of 1,062 people'. What nobody says is the record: that study was retrospective and run with the tablet maker; the randomized trials ran 2,000 FU (370 ads print 2,000). | Chen 2022 never called a trial, no percentages; 'no randomized trial at 10,800' scoped to what PubMed shows. | Slot 03, told straighter than Toku tells it. |
| E | The number on the label, and the metals test proof in every promo send | 287 cocoa ads raise heavy metals, in text; 87 mention a COA or third-party test; 18.2% of cocoa ads lead with a dose. The cocoa playbook's open lane: the COA on video. | No COA, lab result or panel is promised until the team approves it (house rule). | Proof line from the reveal on; video once the COA exists. |
| F | Men's blood flow BFMD territory, not this list's launch | 21.3% of nattokinase ads, 9.0% of cocoa ads and 13.3% of citrulline ads lead with it; Toku runs an ED static to a sexual-health landing page. | Disease-adjacent and a Meta personal-attributes risk; the BFS list is not sold this way. | Recorded, not used on the main list. A BFMD conversation. |
| G | Veterans Day VIP the launch-day occasion (Nov 11) | 5 ads in the census mention veterans. Occasion arcs earn $0.423 per email against $0.219 for bare product launches (content calendar note). | Thank them; never claim a benefit for service-related conditions. | Launch day only; the offer and the VIP lock-in carry it. |
4 · The launch flow
| Slot | Date | Role | Subject (proposed) | Product named |
|---|---|---|---|---|
| 01 | Tue Oct 27 | VALUE 1 · SET THE STAGE | A message for you, {first_name}! / They lied, {first_name}! | no |
| 02 | Thu Oct 29 | VALUE 2 · SECRET REVEALED | Worth knowing, {first_name}! / Let me explain, {first_name}! | no |
| 03 | Tue Nov 3 | VALUE 3 · EDUCATION | Worth reading, {first_name}! | no |
| 04 | Fri Nov 6 | VALUE 4 · PRODUCT ANNOUNCEMENT | This is for you, {first_name}! | yes |
| 05 | Tue Nov 10 | LAUNCH TOMORROW · TELEGRAPH | It's coming, {first_name}! | yes |
| 06 | Wed Nov 11 | LAUNCH · LAUNCH DAY | It's here, {first_name}! | yes |
| 07 | Fri Nov 13 | EXTENDED · EXTENDED / MID | Sorry, {first_name}! | yes |
| 08 | Mon Nov 16 | LAST CHANCE · LAST CHANCE | It's coming to an end, {first_name}! | yes |
Audience on every send: EVERYONE minus the standard suppressions, 11:00 ET. Slot 01 sent Oct 27. Slots 01 to 03 are product-free; the product, its name and the offer arrive at 04; the price and VIP lock-in at 05; launch day is the Veterans Day VIP event.
5 · Recommendation
Sell the morning, not the molecule. The values each reveal one thing this list has not been told, in the order the hub has filed: the two places (01, sent), citrulline and the arginine aisle (02), the dose record nobody tells straight (03). The reveal (04) joins them in the cup the list already drinks, “Japan's breakfast, in the cup you already drink”, with every dose printed and each ingredient doing its own job. The proof in every promo send is the label and, once approved, the metals test.
Three things decide whether this holds. (1) Toku already says the dose line; ours must be the honest version or it reads as a copy. (2) Safety: the bleeding-risk warning must ride on every selling surface, which is also the honest answer to anyone on a thinner. (3) Heat: “warm or cold, never hot” is a promise about a temperature nobody has tested above 50 °C; run the bench test before #4.
6 · Compliance & integrity firewall
| Phrase or surface | Status | Why |
|---|---|---|
| clinically proven | BANNED | clinical DOSES, not trials of this product; NAD made Mars soften 'scientifically proven' to 'studies indicate' on CocoaVia |
| FDA approved | BANNED | false. FDA allows one QUALIFIED claim for high-flavanol cocoa powder (Feb 2023) and requires 'very limited scientific evidence' inside it; our creator videos sa |
| 5 days | BANNED | the '5 days = 2X stem cells' hook is FABRICATED (the 5 was a send delay) and was retired Aug 20 2026; it is still live in the Stem Cells Unlocked e-book, which |
| UC Davis | BANNED | the 2010 cell trial was UCSF (Heiss 2010, PMID 20620742), 16 supervised patients, 30 days |
| Leonard | BANNED | 'Leonard et al., Age, 2013' cannot be located in any database; replace with Sansone 2017 or Heiss 2010 |
| comparable to vigorous aerobic exercise | BANNED | effect-equivalence claim from a press release, not the paper; same shape as 'works like metformin' |
| doubles your stem cells | BANNED | the measure is mobilisation into circulation, not production; cell function did not change in Heiss 2010; no fold figure is verifiable in Sansone 2017 (full tex |
| repairs your heart | BANNED | organ-repair promise from our own creator videos, on the stop list: 'fix your heart, kidneys, liver', 'undo everything', 'keep the transplant table away', 'like |
| clears plaque | BANNED | disease claim; MAIA's 'Destroys Plaque', 'Artery Clean-Up Enzyme', 'clogged arteries' are non-compliant and are not copied |
| dissolves clots | BANNED | drug claim. House framing is 'supports the body's natural breakdown of newly formed clots' with the asterisk, never a thinner |
| lowers blood pressure | BANNED | disease claim; Kim 2008's 5.5 mmHg is a study result in a study population, never the reader's outcome |
| treats heart disease | BANNED | treat / cure / prevent appear only inside the FDA disclaimer |
| Big Pharma | BANNED | house rule: the enemy is named by incentive, never by the literal phrase; doctors are never the villain |
| 36.6% | BANNED | retired by the Jul 23 2026 fact-check: Chen 2022 (PMID 36072877) read as an industry-funded retrospective observational study, '~22% lower carotid IMT, observat |
| hot cocoa | REVIEW | this product is prepared warm or cold, never hot (the enzyme breaks down). The category hook 'You need to be drinking hot cocoa every single morning' cannot be |
| Harvard study | BANNED | Harvard comes through Dr. William Li's credentials, never through a study; COSMOS was 500 mg CocoaVia capsules and its primary endpoint was null (HR 0.90). 'Tha |
| Dr. Lee | REVIEW | the creator-script mangling of Dr. William Li, who is CocoaVia's (Mars) paid partner; crediting a competitor's spokesperson. Li may be cited by name for his cre |
| synergy | BANNED | No published study tests nattokinase with cocoa or with citrulline (PubMed searched Oct 26 2026: 4 and 3 hits, none a study of the pair). Each ingredient is bac |
| they moved the line (unnamed) | REVIEW | An unnamed 'they' holds no one accountable and inverted the chronology (the 2004 panel moved the target; the 2009 UCLA study endorsed it later). Name the panel |
| R1 safety | warning | Label and every PDP/ad landing page need a bleeding-risk caution covering anticoagulant/antiplatelet users (including aspirin), bleeding disorders, and upcoming surgery. Copy must never describe the product as safe alongside blood thinners or 'gentle on the blood'. Do not target or address people who are on blood thinners. |
| R2 safety | blocked | No 'natural alternative to blood thinners', 'aspirin alternative' or 'instead of your meds' framing anywhere (ads, email, SMS, UGC, creator briefs). This is also a disease claim under 21 CFR 101.93(g)(2)(vi) (see R6). Add 'Do not stop or change any prescribed medication' to the caution. |
| R3 safety | warning | FDA expects the label's conditions of use to exclude bleeding-risk subpopulations. The warning cannot be a generic 'consult your doctor'. It should name anticoagulant/antiplatelet use, bleeding disorders, surgery, and pregnancy or nursing. No FDA-mandated nattokinase warning text was found, so counsel must set the exact wording. |
| R4 claims | warning | Do not say 'clinically proven safe' or 'clinically proven dose'. Do not repeat the study's atherosclerosis or plaque outcomes, which are disease claims (R6). At most say 'the amount used in a 12-month human study'. Note: the study recorded no adverse effects but did not exclude aspirin users, and co-administration with aspirin was reported as 'synergetic'. That framing must not appear in marketing. |
| R5 claims | gated | Before launch, get the nattokinase supplier's written regulatory basis: an acknowledged NDI number, or ODI evidence of marketing before Oct 15, 1994, for this form and dose (10,800 FU, uncoated powder). The NDIs found covered other conditions (enteric-coated caplet at 0.01% nattokinase, and a bulk ingredient). Marketing makes no 'FDA-approved', 'GRAS' or 'FDA-notified' claims. |
| R6 claims | blocked | BLOCKED wording: 'dissolves clots', 'clot-buster', 'prevents blood clots/DVT/stroke/heart attack', 'blood thinner', 'thins your blood', 'clears arteries/plaque', 'lowers blood pressure', 'lowers cholesterol', 'for hypertension', 'works with your statin/aspirin'. Pictures of clots, arteries or plaque count too (101.93(g)(2)(iv)(E)). Allowed direction: 'supports healthy circulation' and 'supports normal blood flow', with the DSHEA disclaimer (R9). |
| R7 claims | blocked | Mechanism copy must not describe the product as anti-platelet, anticoagulant or thrombolytic. 'Fibrinolytic enzyme' in educational copy is risky, because the preamble treats drug-mechanism language as implying disease use. Counsel should rule on whether 'supports the body's natural fibrinolytic activity' can be used. |
| R8 claims | blocked | Nattokinase-specific confirmation of R6: no DVT, travel-clot, hypertension or cholesterol framing, including 'borderline' or 'prophylactic' softeners. 'Long flights' and 'sitting all day' creative that implies DVT prevention falls in the same bucket. FDA quoted 'improving Blood Circulation' only within that disease list, so the copy should say 'supports healthy circulation' and avoid 'improves circulation'. |
| R14 claims | gated | If used, the claim must be one of the four wordings verbatim, including the 'very limited' qualifier. Do not paraphrase it as 'supports heart health, backed by FDA', and do not shorten it in ads or SMS. FDA REJECTED the petition's '200 mg daily' intake framing and all dark-chocolate claims, so copy must not pair the claim with a daily flavanol amount. Never write 'FDA-approved claim'. |
| R15 claims | gated | It is unresolved whether a multi-ingredient DIETARY SUPPLEMENT drink mix (with nattokinase and citrulline) can carry this claim. The letter's enforcement-discretion factors are written for conventional foods. Also unresolved: whether the cocoa is 'naturally conserved' at 4% or more, and whether the RACC and nutrient conditions can be met. Default: do NOT use the QHC on label, PDP, ads, email or SMS until counsel signs off. Plain factual content ('1,200 mg cocoa flavanols per scoop') is a separate question for substantiation by CoA. |
| R16 safety | warning | Cocoa is a known lead and cadmium carrier. Ops needs per-serving heavy-metal CoAs for each lot; the binding comparison for lead is 0.5 µg per day of use, at the label's maximum daily servings. If a lot exceeds that, a Prop 65 warning is needed on CA-bound sales (Shopify, Amazon, TikTok Shop). The LII copy of 25705 is labelled '[Operative until 10/1/2025]', so counsel should confirm the current NSRL table. No 'heavy-metal free' or 'lead-free' claims. |
| R17 safety | warning | A second federal benchmark for the CoA review alongside Prop 65 (R16). Lead per daily serving should sit well below 8.8 µg, and the Prop 65 MADL of 0.5 µg is the stricter bar. Keep 'tested for heavy metals' copy strictly factual and backed by lot CoAs. |
| R19 ads | gated | All Meta campaigns: age 18+ targeting. No 'in 30 days' or 'in 2 weeks' outcome promises, no doctor-endorsed cure claims, no body-shaming hooks. Meta blocks scripted fetches, so these quotes came through a model-based extractor as fragments. Re-read the live page before citing them externally. |
| R20 ads | warning | No 'Are you on blood thinners?', 'Worried about your clots?', 'Your blood pressure...' or 'Over 50 with poor circulation?' hooks. Write in the third person or as general statements ('Circulation slows with age' rather than 'Your circulation is slowing'). This also keeps the R2 audience out of the targeting. |
| R21 ads | warning | Drug-like nattokinase framing ('clot-dissolving enzyme', 'natural blood thinner') raises the risk of Meta classifying the product as unsafe or as a drug, on top of the FDA problem. Lead with cocoa and food-style creative. Expect manual review and keep a policy-safe variant ready. |
| R23 claims | gated | Every efficacy claim (circulation, energy, focus, pump, heart) needs a substantiation file of human RCTs on the same ingredient, dose and form. The combination product itself has not been studied, so claims must be ingredient-level and dose-matched: the 10,800 FU study (R4), the cocoa-flavanol trials at matching doses, and citrulline at 2 g. ftc.gov blocks scripts, so the quotes came via an extractor. Re-check before citing externally. |
Formula safety from canon: never attack cocoa, natto, the enzyme, citrulline, watermelon or the morning cup; no synergy; no supplier names, input figures or fill weights; no citrulline dose until the label.
7 · References
- Research dossier: https://bfs-cocoa-natto-research-dossier.vercel.app/ (census, hooks, keywords, competitors, price, regulatory, evidence spine, our list)
- Swipe gallery: https://bfs-cocoa-natto-swipe-gallery.vercel.app/
- product-dev-bot: cocoa-powder playbook (Oct 5), nattokinase playbook (Oct 5), nitric-oxide playbook (Oct 24), cocoa-nattokinase / Toku report (Oct 27)
- Evidence:
_research/reference/slot02/findings.md, canon evidence E1–E33
Built Oct 27 2026 by
_research/build_angles.py. Every number is read from a data file; the recommendation is a recommendation and says so.