Current Agenda

1. Executive Summary

🎚️ The creative got better and the account got worse. CTR went up, CPM stayed flat, and CVR fell off a cliff. By our own model that means the problem is not the ad — it's the offer, the landing page, and an account we rebuilt from zero.

• Last 30 days: $35,573 spent, $20,829 back. 0.59 ROAS, $167 CPA on a $97.79 product.
• v1 did 1.08 ROAS at $79.69 CPA in June, then got disabled. Same product, same team, same ads — half the return after the move.
• Add-to-cart → purchase fell 32% → 17.5%. People are arriving and not buying.

🗺️ We have 116 defined variable values and we've spent real money on about 30. 58% of our budget sits on one angle. 54% sits on one hook style. We've used 5 of 26 subavatars. We are not out of ideas — we've never opened the box.

2. Goals

• CVR 1.45% → 3%+. That's the month. ROAS follows it.
• 1.0 ROAS at $1,500/day by Sept 30 (now ~$1,186/day at 0.59).
• No budget increase until we hold 1.0 for two straight weeks.

3. Targeting

🎯 Targeting is the thinnest section in this agenda, and that is itself the finding. What follows is what the account data actually says; where we have no documented targeting, it says so rather than inventing a strategy.

• Avatar buckets in play: `AV:Bulking` · `AV:OTG` · `AV:Busy`. Only the first two have ever been funded — `AV:Busy` has never run (see §8).
• Age and gender is where the account actually performs, and it is not where the money is:
  • 55–64 M — 0.84 ROAS at $113 CPA. Our most efficient audience, and our least funded.
  • 18–24 M — 0.71 ROAS at $143 CPA. Our best large audience.
• The mismatch: our two most efficient audiences map to subavatars we have never once run (`SA:Sarcopenia Man` for 55–64 M, `SA:Gym Beginner` for 18–24 M). That is the whole thesis of the Subavatar test in §4.
• Not documented here: interest stacks, lookalike or broad settings, placement splits, and any exclusion lists. We cannot describe our targeting because the agenda does not record it. Action: record the targeting on every new test from September on, or we will keep re-learning the same three demographic facts.

4. Portfolio Allocation

🧪 6 tests · 120 concepts. One test per variable, run in ladder order. Every value briefed must come from the approved lists in §8.

Portfolio — fund the best, not the biggest · 12 concepts

• Insight: [Meme Goofy reaction #4](https://projects.motionapp.com/organization/66aca2f9aaf204ba518ef3bc/66aca3148e20c0604ae9476e/creative/6a6dbee1578674433000db7d?startDate=2026-07-15&endDate=2026-08-14&filters=W10=) took $6,430 at 0.80. Clip Header Meme #4 did 1.48 on $233 — 22x less budget to the better asset.
• Hypothesis: our meme problem is budget allocation, not creative. Iteration is supposed to mean copying the winner; we've been copying the loudest.
• Test: re-cut and properly fund the two actual winners. Cap any single creative at 15% of monthly spend.
• Scale if either clears 1.0 at $1,500+. Kill if both fall under 0.7 with real budget.

Subavatar — open the other 21 · 22 concepts

• Insight: we have 26 subavatars and have spent money on 5. Not one of those 5 clears 0.50. Skinny Guy is our most-produced (42 ads) at 0.41, and our second-biggest bucket is `SA:General` — 25 ads to nobody in particular.
• Hypothesis: we don't have a subavatar problem, we have a subavatar range problem. Our two most efficient audiences map to subavatars we have never once run.
• Test: three untested subavatars, picked off our own list to match where the money already works:
  • `SA:Sarcopenia Man` — 55–64 M is our most efficient audience (0.84 ROAS, $113 CPA) and our least funded. This is that person, never tested.
  • `SA:Gym Beginner` — 18–24 M is our best large audience (0.71, $143 CPA).
  • `SA:GLP-1` — promised since June, still $0 spent.
• Written as people, in their own language from reviews and Reddit. Stop defaulting to `SA:General`.
• Scale if any one clears 0.8. Kill if all three land under 0.5 — that points at the offer, not the audience.

Angle — benefits beat features · 24 concepts

• Insight: set High Protein aside and feature angles do 0.27, benefit angles do 0.54. `AN:Complete Nutrition` (0.29) and `AN:Balanced Macros` (0.15) describe the powder. `AN:Stay Satiated` (0.72) describes the customer's life.
• Hypothesis: 58% of our money sits behind `AN:High Protein` — a spec, not a reason, and a beginner claim aimed at an audience that's heard it for years. We have 21 angles available and have tested 9.
• Test: cut Complete Nutrition and Balanced Macros (32 ads, $3,098). Fund `AN:Stay Satiated` properly. Open four untested benefit angles: `More Energy`, `Recover Better`, `Less Overwhelmed`, `Look Better`.
• Scale if benefit-led angles beat the High Protein control.

Format — scale Progress, buy some proof · 20 concepts

• Insight: `FS:Progress` is the best-performing format in the account — 1.13 ROAS at an $87 CPA. It's a static, and it's had $609.
• Hypothesis: progress/journey framing pre-qualifies the buyer. Separately, we have no proof formats running at all — no `FV:Testimonial`, no `FV:Personal Story`, no `FS:Review`, no `FS:UGC Story` — which is the likeliest reason people add to cart and don't buy.
• Test: scale `FS:Progress` across two subavatars. Open the proof formats: `FV:Testimonial`, `FV:Personal Story`, `FS:Review`, `FS:UGC Story`.
• Scale if Progress holds 1.0+ at $3,000 (5x current). Kill if under 0.7 — then 1.13 on $609 was noise.

Hook — put an open loop in it · 26 concepts

• Insight: we hook 23.5% and hold 12.4%. Our best hookers are our worst holders — POV 43.9% → 8.2%, Static-to-Video 44% → 7.2%. Nothing above a 40% hook beats 0.80 ROAS.
• Hypothesis: our hooks have attributes and emotion but no open loop. A meme lands its joke in two seconds — the loop shuts before anyone gets curious. `HS:Meme` is 54% of spend.
• Test: we have 22 hook styles and have run 6. Open the ones that are structurally open loops: `HS:Curiosity Gap` (literally the open-loop hook, never tested), `HS:Contrarian`, `HS:Mistake Callout`, `HS:Objection Callout`. Every concept must survive cut the first 3 seconds — is there still a reason to watch?
• Scale if hold reaches 22% and beats the meme control on ROAS.

Stage of Awareness — rebuild BOFU · 16 concepts

⭐ Last on the ladder, first in priority. This test's result decides whether September is a creative month or an infrastructure month.

• Insight: v1's four best ads (1.11–1.99 ROAS) were all bottom-of-funnel. v2 has run one BOFU ad all month.
• Hypothesis: we didn't lose BOFU performance, we stopped making BOFU. Restoring it recovers the add-to-cart → purchase gap.
• Test: proof-led, most-aware creative — pairs directly with the proof formats above, plus `HS:Social Proof`, `HS:Testimonial`, `HS:Authority`, all untested.
• Scale if BOFU clears 1.0 and ATC → purchase gets back over 25%. Kill if ATC → purchase stays under 20% — that confirms it's the landing page and offer, not the creative.

5. Production

🚨 174 concepts is why we can't read our own account. At $45,000/month that's $259 each — about one and a half sales at our CPA. Last month we ran 200 ads and only 30 got enough spend to rank. September ships 120 at $375 each.

Week 1 fixes — before any of the above matters

• v1 reinstatement or v3 warm-up. v3 has never spent a dollar — it is not a backup, it's an empty account.
• Review the Bulking Listicle landing page and the offer. That's where the CVR collapse points.
• Settle the `FS:` conflict — see §8. Half our funnel-stage history depends on it.
• 34% of last month's spend was untagged and invisible to every number in this doc.
• Check blended ROAS in Shopify/Northbeam before we act on any spend cuts.

6. KPI Targets

⚖️ Where this could be wrong. The best signals are small — Progress is 2 ads on $609, Stay Satiated is $230, the meme winners are $233 and $287. They're leads to fund, not proven winners, which is why every test has a kill line.

>

And the big one: if CVR is really an offer and landing-page problem, this roadmap helps at the margins while the real fix sits elsewhere. The Stage of Awareness kill criterion is built to catch that.

7. Strategist Allocation

• Robert (24) — Angle, Hook, statics. Also owns every Week 1 fix.
• Zubash (20) — Portfolio (pure iteration, not new work), Hook.
• Emylee + Producer (16) — Subavatar, Stage of Awareness shoot assets. No shoot asset ships without full naming.
• Retainer Creator (14) — Format, Hook. Autonomy on execution; the ask is longer, story-led cuts with a loop that stays open.
• UGC roster (22) — Format, Subavatar, Angle.
• Reach Digital (24) — Stage of Awareness, Angle, Subavatar. Scoped briefs with kill criteria attached. Continues owning naming enforcement.

8. Approved Variables

⚠️ `FS:` conflict — needs a decision in week 1. This list defines `FS:` as Creative Format – Static. But every ad currently running uses `FS:` for funnel stage (TOFU / MOFU / BOFU), and that's what the BOFU finding above is built on. Two options: give static format its own key and leave `FS:` as funnel stage, or move funnel stage to a new key. Either way, do not brief `FS:` until this is settled or we lose the ability to compare September to anything before it.

Coverage — how much of our own taxonomy we actually use

Avatar `AV:` — Bulking · OTG · Busy Subavatar `SA:` — Skinny Guy · Skinny Girl · GLP-1 · Skinnyfat · Gymbro · Glute Girl · Gym girl · Gym Beginner · Post-Menopausal Woman · Sarcopenia Man · Bodybulder · Medical Condition · Hybrid Athelete · Endurance Athlete · Casual Hiker · Travel Backpacker · Casual Runner · Elite Endurance Athlete · Busy Office Worker · Busy Blue Collar Worker · Busy Parents · Breakfast · Out & About · No Break Workers · ADHD · General Angle `AN:` — Convenient · Healthy · High Protein · Balanced Macros · Complete Nutrition · Be More Consistent · Save Time · Save Money · Save Stress · Tasty · Reach Goals Faster · Self-Control · Less Overwhelmed · More Energy · Stay Satiated · More Free · Recover Better · Lock in More · Look Better · Breakfast Hack · Ingredients Hook Style `HS:` — Reaction · POV · Trend · Transformation · Listicle · Contrarian · Curiosity Gap · Problem Callout · Avatar Callout · Feature/Benefit Callout · Symptom Callout · Authority · Shocking · Disgusting · Hidden Secret · Desired State Callout · Mistake Callout · Social Proof · Testimonial · Meme · Before & After · Objection Callout Creative Format – Static `FS:` — Listicle · Longform Copy · Ugly Ad · UI · Review · Carousel · Offer · UGC Story · Advertorial · Avatar Callout · Symptom Callout · Benefit Callout · Feature Callout · Pubity · Us vs Them · Meme · Text Wall · Post-It · AMA · Progress Creative Format – Video `FV:` — Listicle · Ugly Ad · Caption Only · Motion Graphics · Symptom Callout · Avatar Callout · Benefit Callout · Feature Callout · Transformation · Warehouse · Skit · Voiceover Narrative · Personal Story · Testimonial · Reaction · Taste Test · Street Interview · Podcast · Mashup · Us vs Them · Text Overlay · Side by Side · Comparison · Educational