The strobe-cut before/after works when the caption is a dismissal she has actually heard. The age variants dismissed her age. These four dismiss the problem she came in with: the belly, the back, the floor, the wobble.
Two sources disagree in a useful way. The research page says what happens to her body. The quiz says what she arrives asking about.
Two cautions on the numbers. These respondents were recruited by the body-transformation ad, so the weight goal is inflated by the creative that brought them. It is still the dominant felt problem by a mile. Balance is the mirror image: lowest felt demand, strongest evidence. The research says it should anchor the product; the quiz says almost nobody arrives asking for it. That argues for building belly first and aiming balance at the 60+ bracket, where the CTR is already highest.
Left out of this set: sleep and daytime energy (58% and 54%) are real but not a physical before/after, so they become after-contexts. Leaking (76% at least occasionally) is the biggest hidden issue and the wrong thing to stage in a public before/after. Hot flushes stay off the table entirely.
A per-person crosstab of the same 60 days, 473 women who answered the goal question.
| Picks "lose weight around my middle" | 45–49 | 50–54 | 55–59 | 60–64 | 65+ |
|---|---|---|---|---|---|
| Share of that age bracket | 63% | 87% | 79% | 90% | 79% |
| Same women, what they live with | Weight goal (n=339) | Any other goal (n=69) |
|---|---|---|
| Stiff in the mornings | 48% | 52% |
| Energy dips through the day | 55% | 56% |
| Broken sleep | 88% | 83% |
| Cannot get off the floor unaided | 47% | 59% |
| Lower back stiff most mornings | 57% | 57% |
| Stairs are hard | 52% | 65% |
| Leaks at least sometimes | 75% | 83% |
| Fails the one-leg test | 33% | 54% |
| Shape changed around the middle | 69% | 49% |
Read. The middle is the ticket she buys with, not the problem she lives with. Four in five name it as the goal at every age and every stage, yet the women who name it carry exactly the same stack of stiffness, broken sleep, flat afternoons, a back that seizes and a floor they avoid as the women who don't. Only balance separates the groups. Three reasons to discount the 82% without dismissing it: the goal question forces one answer, the weight option sits first and is the only appearance option, and this sample was recruited by a body-transformation ad. Why it matters to her is unambiguous: feel better in myself 82%, stay independent 72%, heart and bones 70%.
For the funnels: the dismissal in the hook can name the middle, but the before beats should show the stack she actually has, the couch, the floor, the stairs, the back, tired by three, and the after must show ease and energy rather than a smaller body. Worth testing in the quiz itself: rotate the goal options, or add "and what else?" as a second pick.
The dissection found four load-bearing elements. Everything else the source ad had was dropped without losing performance.
Before: profile, hands under the belly.
After: same profile, same head height, hands on the waist, ribcage lifted.
Same locked-off framing. The delta is the silhouette between the hands.
Before: seated on the bed edge, hand on the back.
After: seated on the same bed edge, upright, hands loose on the thighs.
Seated pairs keep the head still; the after then stands in one movement.
Before: sat on the carpet, legs to one side, propped on an arm.
After: sat on the same carpet, cross-legged, upright, hands on knees.
Floor-seated pairs match head height; the after then rises with no hands.
Before: standing in the doorway, one hand gripping the frame.
After: standing in the same doorway, one foot lifted, hands free.
Standing pairs work here because both are upright at the same mark.
Two deep-research runs: 6,251 Reddit posts on stiffness and mobility, and 316 threads plus 1,080 YouTube comments on desires and fears. They surface physical symptoms the quiz never asks about.
| Symptom, in her words | Signal | Before/after potential | Claim |
|---|---|---|---|
| After-sitting stiffness. "I get up like I'm 95", "Tin Man", "worse after sitting or driving" | The most repeated physical complaint; felt as surprising. Not in the quiz. | Best new before beat. Off the couch or out of the car in stages. Upgrades B and C; the car after-context becomes proof. | Moving easily. Claim carefully. |
| Hands. "Finger joints so swollen I can't wear rings", "got my thumbs back and could open jars" | Named first in her joint list. Quiz: 16%. | The jar lid. Cheapest, most recognisable before/after in the set. Candidate fifth cell. | Grip strength only. Joint pain is HRT's claim in her mind. |
| Stairs. Knee pain going down, "out of breath going up", "climb stairs without being out of breath" | Early, specific pain signal. Quiz: 43% puffed at the top. | Hand on the rail, one step at a time, becomes two at a time. After 2 for C and D. | Strong. The loaded stair climb is the programme's retest. |
| The floor, gone. "I haven't touched the floor for five years", "I avoid sitting because it's so hard to get back up" | Flagship functional fear. Quiz: 51%. | Reinforces C and hands it a hook line. | Strong. |
| Frozen shoulder. "The first sign" | Widely described. Quiz: 39% shoulders or neck. | Top shelf, coat sleeve, bra strap. Good visual. | Amber. Peri versus pre not significant; physio strengthening made it worse for some. |
| Feet and ankles first thing. "My ankles are toast", plantar fascia, Achilles | Common. Quiz: 30%. | Hobbling first steps out of bed. A before beat, not a funnel. | Weak. |
| Night pain, elbows. 3am hips "crushed in a vice"; "sudden onset tennis elbow, both arms" | Common; surprisingly often. | Not visual, or too niche. | None. |
| Where the weight lands. "Christmas hams down the back of my pants", "lost my edges", "deflated balloon" | 12+ threads. Quiz: 65%. | Vocabulary for A, not a new funnel. | Do not claim. |
| Brain fog. "Losing my mind" | 14+ threads, rivals body distress. Not in the quiz. | Not a physical before/after. A separate whitespace test. | Amber wording only. |
Three warnings the corpus adds to the storyboards. Joint-pain relief belongs to HRT in her mind ("estrogen was the only thing that helped my joints"; a minority say exercise made it worse), so B, hands and shoulder promise moving easily, never pain relief. Her proof grammar is forensic: same outfit, pose and place, and big deltas read as fake or Ozempic after 2024. That cuts against the dramatic-delta rule; the shipped ad's numbers are the only real data, so keep the rule but let the action delta carry more of the load than the body delta. And Meta bans weight-loss before/after imagery outright while the ASA banned a 12-week menopause programme for the belly claims, which makes A the highest-rejection-risk cell and the action-based befores safer on both counts.
Additions: after-sitting stiffness as the first before beat in B and C; stairs as After 2 in C and D; a fifth cheap candidate, E · Hands, the jar, to test alongside.
Two phases, so the creative gets judged before the quiz is touched.
| Phase | What changes | What it answers |
|---|---|---|
| 1 · creative only | Four ads in the existing Lead campaign, one ad set, women 45–65 (60–65+ for balance if split later). Same quiz, split by utm_content=issue-belly | issue-back | issue-floor | issue-balance. | Does a problem-dismissal beat an age-dismissal on CPL, against the A56 control in the same auction. |
| 2 · matched intro | For any winning cell: an issue-matched line on the intro screen and the matching goal preselected on the first goal question. The stiffness, balance and clean-scan interstitials already exist. | Does continuity from ad to first screen lift completion and checkout for that issue. |
Decision rule, carried over from the age variants: run each cell to about 50 leads. Compare cost per lead first, checkout rate second. Kill a cell whose CPL runs 50% above the control; swap in the backup hook before abandoning the issue.
Production is the V40 pipeline unchanged: after sheet, before sheet derived from it, strobe-pair frames, the 50% alpha-blend gate, four clips via minimax_h3, the existing ffmpeg cut list and audio bed. Around 250–350 credits per variant, with the before expected to need one re-roll. The 6-credit de-risk per variant (sheets plus strobe pair plus blend check) answers "is the before convincing" before any video is bought.