Underdays
UK + US
Meta Ads
What Happened — 7–13 May 2026
Sprint Performance Report · Generated 14 May 2026
5 Winners
8 Learning Zone
3 Losers
134 Unclassified (<£400)
£31,050
7-Day Spend
2.00x
Blended ROAS
150
Total Ads
2.6%
Blended CTR
£17.61
CPM
~26%
Avg Thumbstop
🇬🇧 United Kingdom
67% win rate
2.25x
ROAS
£10,791
Spend
68
Ads
4W
1E
1L
62U
🇺🇸 United States
14% win rate
1.95x
ROAS
£17,478
Spend
54
Ads
1W
7E
2L
44U
⚠️ Taxonomy gap: 28 ads (£2,781 spend, ROAS ~1.06x) carry no taxonomy tags and cannot be classified by source, concept, or audience — they appear to use a non-standard naming convention. Excluded from pattern analysis but likely drag blended ROAS. Recommend enforcing naming taxonomy on all active ads.
Sprint Concepts
6 concepts · click to expand
exploit
high
▼
C250 MASHUP SCALE-OUT — Push the podcast-hook VSL to US parity with UK
C250 is the account's highest-spend confirmed winner outside the ecom cluster. In UK at £2,032 it clears the 2.0x threshold at 2.28x (27% thumbstop, 5% p50). The same creative in US across two ad IDs (£988 at 2.18x; £1,576 at 2.08x) is landing just below the tier thresholds. The US versions are structurally strong — 24–31% thumbstop and 5% p50 mirror UK — suggesting the conversion gap is either a landing page localisation issue (GBP vs USD copy mismatch) or a bid/audience temperature problem rather than a creative failure.
Hypothesis
Localising the C250 VSL body copy and CTA to USD pricing and US cultural norms will lift US ROAS above 2.2x at £1,000+ spend because the hook metrics (TS 24–31%) confirm attention is captured but the body is failing to convert.
Hook Guidance
Keep the existing podcast-style opening exactly — it is producing 24–31% thumbstop in both markets. Do NOT change the first 3 seconds. The edit starts at 4–10s: replace GBP price references with USD, swap UK-specific social proof for US equivalents, and ensure the CTA says "Shop now — free US shipping."
Deliverables
2 × mashup/vsl · US variant with USD copy and US LP · everyday product · mixed type · mixed colour · TOF · healthfirst audience
Success Threshold
2.2x ROAS at £1,000+ spend (£1k–£2k tier threshold)
Evidence
[A4] £2,032 · 2.28x ROAS · 2.5% CTR · 27% TS · p50=5% · TOF · uk · everyday | [A8] £988 · 2.18x · 3.2% CTR · 24% TS · p50=5% · TOF · us | [A10] £1,576 · 2.08x · 3.4% CTR · 31% TS · p50=5% · TOF · us | [P1: mashup/vsl/podcast is the only source/concept/hook combination with a confirmed winner above £2k spend.]
Variants
US USD Copy — Identical to A4 video; replace all price references with USD equivalents; swap destination LP to US storefront. Tests whether currency/localisation is causing the US gap. Hook: podcast · CB: underwearskincare.
US Social Proof Refresh — Keep A4 hook and structure; swap mid-body testimonial for US-facing version (US press mention or US-specific stat). Tests whether social proof localisation lifts mid-funnel conversion. Hook: podcast · CB: underwearskincare.
Do
- Keep the podcast visual frame intact (A4 proof)
- Localise body copy for USD and US tone
- Route US ads to a US-optimised LP
- Test at £400 minimum before scaling
Don't
- Change the hook (TS already 24–31% across all 3 ads)
- Add priceoffer hook to C250 (podcast is its identity)
- Run US and UK in the same ad set
- Scale past £1,000 before clearing 2.0x
exploit
high
▼
CUTOUT GRAPHIC EXPLOIT — Build on the account's most efficient qualified combination
A3 (cutout/graphic/priceoffer, blackset, UK) posted 2.63x ROAS at £434 — the highest ROAS of any qualified static ad this sprint. Two unclassified cutout ads show similar efficiency (2.92x blended across £965 total). Despite this, cutout source is only £965 total spend — the account is not exploiting its best-performing format. The graphic concept blended is 2.61x ROAS across £3,013 — highest efficiency concept with >£500 spend.
Hypothesis
Deploying additional cutout/graphic/priceoffer creatives against everyday and starterkit products in both UK and US will produce ROAS ≥ 2.5x at £400+ spend because the format's conversion efficiency is proven at account level and the priceoffer hook is the strongest hook for the trendfollower audience at qualifying spend.
Hook Guidance
Open on a clean cutout product shot — white background, no lifestyle context. On-screen text in the first 2 seconds: the price point prominently displayed ("From £X"). Do NOT open with lifestyle imagery — A3's 0.9% CTR proves this format wins on conversion intent, not scroll-stopping rate.
Deliverables
3 × cutout/graphic · UK everyday/thong/core · US everyday/thong/core · UK starterkit/mixed/core · all priceoffer hook · trendfollower TA · TOF · comfortsimplicity CB
Success Threshold
2.5x ROAS at £400+ spend
Variants
UK Everyday Cutout Price — Same format as A3 but against everyday thong/core. Tests whether the blackset win is product-specific or format-wide. Hook: priceoffer · CB: comfortsimplicity.
US Everyday Cutout Price — US adaptation with USD pricing. Tests US market transferability. Hook: priceoffer · CB: newdrop.
Starterkit Cutout Price — Against starterkit product (proven winner in video). Tests whether price-first static can amplify the starterkit signal. Hook: priceoffer · CB: resetdetox.
Do
- White/clean background (cutout format convention)
- Show price prominently in frame 1
- Use core colourways (not limitededition)
- Deploy UK before US — UK win rate is 4× better
Don't
- Add lifestyle imagery (cutout works by removing noise)
- Apply hype hook — it's 1.55x in this account
- Route to campaign LP without the specific product shown
exploit
high
▼
ORGANIC156 STARTERKIT SCALE — Expand the aestheticDR teaserhook winner into US at scale
A1 (Organic156, UK, aestheticDR, teaserhook, starterkit) is the account's highest ROAS qualified winner at 3.24x on £423 spend. The same creative in US (A7) posted 2.26x at £642 — confirming the concept travels but with a 98-basis-point ROAS gap. The teaserhook category is performing at 2.50x blended (W:1, £1,405) — second-highest hook category ROAS among those with qualifying spend.
Hypothesis
Scaling Organic156 in UK to £1,000+ spend and deploying a US-localised version with USD pricing will lift US ROAS above 2.2x at £1,000 spend because the conversion mechanism (minimalistcomfort audience + resetdetox benefit + teaserhook) has proven demand signal in both markets.
Success Threshold
UK: 2.5x ROAS at £800+ · US: 2.2x ROAS at £1,000+
Evidence
[A1] £423 · 3.24x · 18% TS · p50=5% · uk | [A7] £642 · 2.26x · 22% TS · p50=5% · us
Do
- Prioritise UK scaling first — proven 3.24x
- Use mixed colour/type for broader catalogue appeal
- Route US to starterkit-specific LP
Don't
- Change the teaserhook opening
- Combine with priceoffer (different audience intent)
- Put in MOF campaign — cold-traffic winner only
exploit
medium
▼
UGC NEGATIVEHOOK FITCONFIDENCE — Scale Natalie Gold into UK highwaisted range
A2 (Natalie Gold, ugc/negativehook/vsl, UK, everyday/highwaisted) posted 2.81x at £651 — third-highest ROAS winner. Thumbstop is 30% — highest of any video winner — and p50 is 9%. The negativehook category produces 2.08x blended (W:1, £1,404) with no losers. There is no US equivalent of this creative and no other ugc/fitconfidence/negativehook combination in the qualified set.
Hook Guidance
Open with the negative problem statement in the first 2 seconds — spoken or on-screen: "Not [positive state]? Here's why..." Do NOT open with a product reveal or lifestyle shot. The 30% thumbstop on A2 is the highest video thumbstop in the account — it is driven by the negativehook's pattern interrupt.
Deliverables
2 × ugc/vsl · 1 × US version of A2 negativehook/fitconfidence/everyday highwaisted · 1 × UK extension brief or thong type
Success Threshold
2.5x ROAS at £400+ spend
Do
- Lead with the problem state in spoken audio
- Keep creator-led (ugc format)
- Target fitconfidence explicitly
Don't
- Start with a product shot
- Combine with hype hook
- Run in MOF — this is cold traffic creative
iterate
high
▼
WORKOUT PRODUCT RESET — Kill hype hook on Slate, test teaserhook/negativehook
The Slate workout limitededition launch has produced two losers: L2 (26W17_Meet_Slate, UK MOF, hype hook, 0.77x at £483, PAUSED) and L1 (26W19_C155_BundleLP, US TOF, hype hook, 1.28x at £555, PAUSED). Both use the hype visual hook — the worst-performing hook in the account at 1.55x blended ROAS with W:0 and L:2 across £3,718 spend. A third Slate static (26W18_USP_Slate, priceoffer) is in learning zone at £1,836/1.95x. The workout product may be commercially viable but the creative strategy is structurally wrong.
Hook Guidance
Do NOT open with energy/hype (fast cuts, loud music, product in motion). Instead: (a) teaserhook — open with Slate colourway partially revealed in 3 seconds ("You haven't seen this yet…") or (b) negativehook — "Every workout thong gives you this problem. Except one." Spoken line must land within 2 seconds. Model on A1 (teaserhook) or A2 (negativehook).
Deliverables
2 × ecom/productshowcase · UK Slate teaserhook TOF · US Slate negativehook TOF
Success Threshold
2.5x ROAS at £400+ spend (new hook approach)
Do
- Move Slate to TOF only — L2's MOF placement is a confound
- Change hook first (cheapest variable)
- Test teaserhook before ordering new production
Don't
- Revive L1 or L2 without a new hook (Zombie protocol)
- Use hype hook on any new workout creative (1.55x blended)
- Route to BundleLP destination (L1: bundle LP + hype = 1.28x)
iterate
high
▼
US ECOM EFFICIENCY FIX — Replace emphasizesolution hook on 25W50_C44 with priceoffer
The ecom/productshowcase cluster is the largest single spend block at £11,387. The US fitconfidence everyday/highwaisted ecom (25W50_C44, A13: £2,715, 1.75x) is dragging — 25 basis points below its 2.0x tier threshold. A13 has been running since at least W50 2025 and is showing classic fatigue signals. The emphasizesolution hook is producing 1.82x blended vs priceoffer's 2.35x — a 53-basis-point gap on the same source/concept combination.
Hypothesis
Replacing the emphasizesolution hook on A13 with a priceoffer hook will lift ROAS above 2.0x at £2,000+ spend because priceoffer is the account's most proven hook at this spend tier (A5: £3,534, 2.19x winner) and is 53bps more efficient on the same source/concept combination.
Evidence
[A13] £2,715 · 1.75x · 3.2% CTR · emphasizesolution hook | [A5] £3,534 · 2.19x · 3.5% CTR · priceoffer hook — same ecom/productshowcase source, winner.
Success Threshold
2.0x ROAS at £2,000+ spend
Do
- Test priceoffer hook first — cheapest variable
- Match format exactly to A5 (the proven winner)
- Monitor frequency closely (A13 is oldest active ad)
Don't
- Add new creative complexity — the issue is the hook
- Scale past £3,000 before clearing 2.0x threshold
Creative Patterns
7 patterns · min. 2 ads per sidePodcast visual hook on mashup/VSL consistently outperforms at scale
The C250 mashup VSL family (A4/A8/A10) is the account's highest-spend video concept with a confirmed winner. All three instances show 24–31% thumbstop and 5% p50 — consistent mechanics regardless of market. UK instance clears 2.0x tier at 2.28x; US instances are 2.08–2.18x in learning zone. Creative teams should treat the podcast hook as the default frame for health-benefit VSL content.
Ads: "26W16_C250_V_2 UK mashup" (120245506587130365) · "26W16_C250_V_2 US copy 1" (120245506609590365) · "26W16_C250_V_2 US copy 2" (120245541741130365)
Hype visual hook is the account's most reliable loser signal — W:0 L:2 at 1.55x ROAS
Hype hook ads post 1.55x blended ROAS across £3,718 spend with W:0 and L:2. Both qualified losers (L1: £555/1.28x; L2: £483/0.77x) use hype. The only positive data point is an advertorial in learning zone at 2.35x — still below the 2.5x winner threshold. No hype hook ad has cleared the winner threshold at qualifying spend this sprint. Hype hook is demonstrably failing on new product launches.
Ads: "26W19_C155_BundleLP" (120246768330290365) · "26W17_Meet_Slate_S_1" (120246066517430365) · "26W18_C155_S_1" (120246070731300365)
Priceoffer hook is the strongest hook by ROAS at high spend — W:2 at 2.35x blended
Priceoffer generates 2.35x blended ROAS across £7,653 spend with W:2, E:1, L:1. A5 (£3,534, 2.19x at the 2k+ tier) is the account's highest-spend non-Slate static and its US anchor. The L3 priceoffer loser (£469, 1.99x) is borderline — 1 basis point below threshold, currently PAUSED. Priceoffer is the right hook for product-first ecom/productshowcase content for the trendfollower audience.
Ads: "26W15_USP_White_S_1" (120245022091910365) · "26W17_C201_S_3" (120245761047860365) · "26W11_USP_PolkaDot_S_3" (120243890820630365)
Workout product (limitededition/Slate) is failing across all creative approaches — W:0 at 1.79x blended
Workout product posts 1.79x blended ROAS across £5,810 spend with W:0, L:2. The limitededition Slate colourway has two paused losers and one learning-zone ad still below threshold. The everyday product by contrast is at 2.08x with W:3 at £19,580. The workout product may need either a different benefit angle (underwearskincare or comfortsimplicity — the everyday winners) or a complete hook strategy reset before further investment.
Ads: "26W17_Meet_Slate_S_1" (120246066517430365) · "26W19_C155_BundleLP" (120246768330290365) · "26W18_USP_Slate_S_1" (120246408751090365)
UK consistently outperforms US on efficiency — 2.25x vs 1.95x, 67% vs 14% winner rate
UK produces 4 of 5 winners while holding 35% of spend. US holds 56% of spend but only 1 winner. This is not a creative quality gap — C250 has 24–31% thumbstop in both markets — but likely reflects a post-click conversion gap (landing page, USD pricing, checkout experience). Every US video winner has a direct UK equivalent outperforming it by 20–50 basis points.
Ads: "Organic156 UK" (120245971143630365) at 3.24x vs "Organic156 US" (120245340222480365) at 2.26x · "C250 UK" (120245506587130365) at 2.28x vs "C250 US" (120245541741130365) at 2.08x
Graphic/cutout static with low CTR (0.9%) can produce top-tier ROAS (2.63x)
A3 posted 0.9% CTR but 2.63x ROAS at £434 — third-highest ROAS winner. The graphic concept blended is 2.61x. Low CTR on a static ad is not a failure signal when the priceoffer hook is pre-filtering for purchase-intent visitors. Creative teams should not suppress static ads on CTR alone; the conversion rate per click is the diagnostic that matters.
Ads: "26W17_C201_S_3_s-cutout" (120245761047860365)
Advertorial concept underperforming vs product's VSL peers — W:0 at 1.65x blended
Advertorial concept produces 1.65x blended ROAS across £1,942 spend. The loser advertorial (L1: £555, 1.28x, BundleLP destination) confirms that advertorial content routed to a bundle landing page is particularly weak. The concept may work on its own LP but is failing when pointed at product-collection destinations.
Ads: "26W18_C155_S_1" (120246070731300365) · "26W19_C155_BundleLP_S_1" (120246768330290365)
What Makes GREAT
3 characteristics · 2+ winner ads eachStrong thumbstop (≥20%) on video ads combined with purchase-intent hook produces winner-tier ROAS
The three video winners (A1/A2/A4) all post thumbstop ≥18% and use hooks that pre-filter for audience intent (teaserhook, negativehook, podcast). By contrast, emphasizesolution videos have comparable thumbstop but fail on conversion — suggesting that stopping the scroll is necessary but the hook type determines whether the stopper intends to buy. High thumbstop + wrong hook = high CPM waste.
✦ "26W15_C215_WL_LP Natalie Gold" (120245202208830365) — 30% thumbstop · 2.81x ROAS ✦ "26W16_C250_V_2 UK" (120245506587130365) — 27% thumbstop · 2.28x ROAS
Product-audience fit specificity — pairing precise audience identity with product's primary functional claim
A1 (minimalistcomfort + starterkit + resetdetox) and A2 (fitconfidence + everyday/highwaisted + comfortandstyle) both pair a precise audience identity with the product's core job-to-be-done. They don't try to be everything. This specificity drives higher conversion efficiency vs the broad healthfirst/underwearskincare pairing (1.88x on £8,230 spend).
✦ "26W13_Organic156 UK" (120245971143630365) — minimalistcomfort/resetdetox — 3.24x ✦ "26W15_C215_WL_LP" (120245202208830365) — fitconfidence/comfortandstyle — 2.81x
UK-first creative testing before US scale preserves budget efficiency
All 4 UK winners entered at £400–£650 spend and hit threshold before being deployed at scale. The US losers (L1/L2) and the struggling US ecom cluster were deployed at spend levels before creative confirmation. UK is the right proving ground for new creative concepts — 67% winner rate among qualified UK ads vs 14% for US.
✦ "26W13_Organic156 UK" (120245971143630365) — £423/3.24x UK winner before US test ✦ "26W17_C201 cutout UK" (120245761047860365) — £434/2.63x UK winner with no US equivalent yet
How to Fix GOOD → GREAT
3 learning-zone fixesA7 (Organic156 US) is at 2.26x (£400–999 tier — threshold 2.5x) — post-click US localisation gap
A7 is the US mirror of A1 (3.24x UK winner) with near-identical thumbstop (22% vs 18%) and p50 (5%) but 98 basis points lower ROAS. The creative mechanics are working — the gap is almost certainly USD pricing and/or UK-specific copy in the body. A single LP and copy localisation pass (USD, US shipping, US return policy) should close a substantial portion of this gap and push A7 toward the 2.5x winner threshold.
Ads: "26W13_Organic156_V_1_r-us" (120245340222480365) · "26W13_Organic156_V_1_r-uk" (120245971143630365)
Ads: "26W13_Organic156_V_1_r-us" (120245340222480365) · "26W13_Organic156_V_1_r-uk" (120245971143630365)
26W16_C250 US mashup family is at 2.08–2.18x (£400–£1,999 tiers — thresholds 2.5x/2.2x) — US conversion localisation
Three instances of C250 are in the US learning zone with consistent thumbstop and p50. The structural creative is confirmed working (UK winner at 2.28x). Localising the body copy and CTA to USD pricing and US cultural norms should push the £988 instance to 2.5x and the £1,576 instance toward 2.2x, unlocking two new winners from existing creative spend.
Ads: "26W16_C250 US £988" (120245506609590365) · "26W16_C250 US £1,576" (120245541741130365)
Ads: "26W16_C250 US £988" (120245506609590365) · "26W16_C250 US £1,576" (120245541741130365)
A12 (26W18_USP_Slate US) is at 1.95x (£1,000–£1,999 tier — threshold 2.2x) — hook strategy reset needed
A12 is the Slate workout launch's best performer (£1,836, 1.95x) using the priceoffer hook, 25 basis points below threshold. Unlike the hype-hook Slate losers, this ad has the right hook type — but the workout/limitededition product has no precedent of qualifying at this ROAS level. A teaserhook iteration (proven at 2.50x account-wide) on the same Slate product brief may unlock the 25-basis-point gap needed to clear 2.2x.
Ad: "26W18_USP_Slate_S_1_r-us" (120246408751090365)
Ad: "26W18_USP_Slate_S_1_r-us" (120246408751090365)
Performance Breakdowns
By Source
| Source | Ads | W/E/L | ROAS | Spend |
|---|---|---|---|---|
| ecom | 37 | 1/3/3 | 1.97x | £13,818 |
| mashup | 24 | 1/2/0 | 1.85x | £6,692 |
| aestheticDR | 22 | 1/1/0 | 2.16x | £2,642 |
| aestheticLF | 12 | 0/2/0 | 2.53x | £2,212 |
| ugc | 15 | 1/0/0 | 2.29x | £1,419 |
| cutout | 5 | 1/0/0 | 2.92x | £965 |
By Concept
| Concept | Ads | W/E/L | ROAS | Avg TS |
|---|---|---|---|---|
| productshowcase | 26 | 1/2/2 | 2.04x | 7% |
| vsl | 55 | 3/3/0 | 1.98x | 30% |
| graphic | 20 | 1/1/0 | 2.61x | 5% |
| advertorial | 10 | 0/1/1 | 1.65x | 0% |
| ugc | 3 | 0/1/0 | 1.75x | 20% |
By Visual Hook
| Hook | W/E/L | ROAS | Spend |
|---|---|---|---|
| priceoffer | 2/1/1 | 2.35x | £7,653 |
| podcast | 1/2/0 | 1.89x | £6,488 |
| emphasizesolution | 0/2/0 | 1.82x | £4,551 |
| hype | 0/1/2 | 1.55x | £3,718 |
| expertvoice | 0/1/0 | 1.95x | £1,502 |
| teaserhook | 1/1/0 | 2.50x | £1,405 |
| negativehook | 1/0/0 | 2.08x | £1,404 |
| satisfyingintro* | –/–/– | 3.19x | £798 |
| dramatisetheproblem* | –/–/– | 2.60x | £637 |
* All under £400 (unclassified) — potential signals worth watching
By Target Audience
| Audience | W/E/L | ROAS | Spend |
|---|---|---|---|
| healthfirst | 1/4/0 | 1.93x | £9,882 |
| trendfollower | 2/1/2 | 2.18x | £8,587 |
| fitconfidence | 1/1/0 | 1.94x | £4,248 |
| everydayathlete | 0/1/1 | 1.70x | £2,638 |
| minimalistcomfort | 1/1/0 | 2.35x | £2,432 |
| wellnessaesthetic* | –/–/– | 3.64x | £462 |
* All under £400 — strong unclassified signal
By Product
| Product | W/E/L | ROAS | Spend |
|---|---|---|---|
| everyday | 3/5/1 | 2.08x | £19,580 |
| workout | 0/2/2 | 1.79x | £5,810 |
| starterkit | 1/1/0 | 2.13x | £1,818 |
| blackset | 1/0/0 | 2.92x | £965 |
By Funnel Stage
| Stage | Ads | W/E/L | ROAS | Spend |
|---|---|---|---|---|
| TOF | 113 | 5/8/2 | 2.05x | £26,615 |
| MOF | 9 | 0/0/1 | 2.18x (audience temp ≠ creative quality) | £1,654 |
| BOF | 0 | – | – | – |
By Core Benefit
| Benefit | W/E/L | ROAS | Spend |
|---|---|---|---|
| underwearskincare | 1/3/0 | 1.88x | £8,230 |
| newdrop | 1/1/2 | 2.16x | £7,656 |
| comfortsimplicity | 1/1/0 | 2.05x | £4,569 |
| resetdetox | 1/2/0 | 2.01x | £3,170 |
| performance | 0/1/1 | 1.79x | £2,459 |
| comfortandstyle | 1/0/0 | 2.45x | £849 |
Budget Split
45% Exploit
35% Iterate
20% Explore
The winner pool is thin (5 qualified winners) but concentrated in high-conviction concepts (C250, Organic156, cutout/graphic, Natalie Gold). The ecom/productshowcase US cluster needs structural fixes before further scaling. The Slate workout launch requires a full hook reset before reactivation. Explore budget should target: satisfyingintro and dramatisetheproblem hooks (both 2.6–3.2x ROAS in unclassified territory) and the wellnessaesthetic TA (3.64x unclassified, no qualified data yet).
Top Ads — GREAT + GOOD
Top 10 by ROAS at £400+ spend| Status | Ad | Spend | ROAS | CTR | Thumbstop | p50 | Tags |
|---|---|---|---|---|---|---|---|
| Winner |
26W13_Organic156_V_1 · aestheticDR · uk · starterkit
120245971143630365
Top ROAS winner; teaserhook + resetdetox on starterkit entry product
|
£423 | 3.24x tier 2.5x |
2.5% | 18% | 5% | aestheticDRvslminimalistcomfortTOFuk |
| Winner |
26W15_C215_WL_LP_V_2 · ugc · natalie gold · uk
120245202208830365
Highest thumbstop in account (30%); negativehook + highwaisted type
|
£651 | 2.81x tier 2.5x |
3.1% | 30% | 9% | ugcvslfitconfidencenegativehookuk |
| Winner |
26W17_C201_S_3 · cutout · uk · blackset
120245761047860365
0.9% CTR but 2.63x ROAS — priceoffer filtering purchase-intent traffic
|
£434 | 2.63x tier 2.5x |
0.9% | — | — | cutoutgraphictrendfollowerpriceofferuk |
| Winner |
26W16_C250_V_2 mashup UK · everyday
120245506587130365
Highest-spend video winner; podcast hook holding above 2.0x threshold at scale
|
£2,032 | 2.28x tier 2.0x |
2.5% | 27% | 5% | mashupvslhealthfirstpodcastuk |
| Winner |
26W15_USP_White_S_1 · ecom · us · everyday
120245022091910365
Highest-spend winner overall; priceoffer/ecom anchor for US market
|
£3,534 | 2.19x tier 2.0x |
3.5% | — | — | ecomproductshowcasetrendfollowerpriceofferus |
| Learning |
26W13_Organic156_V_1 · aestheticDR · us · starterkit
120245340222480365
US mirror of UK winner; 98bps gap likely LP/copy localisation issue
|
£642 | 2.26x needs 2.5x |
3.5% | 22% | 5% | aestheticDRvslminimalistcomfortus |
| Learning |
26W18_C155_S_1 · ecom · us · workout advertorial
120246070731300365
Best workout ad this sprint; closest to threshold on Slate/performance angle
|
£531 | 2.35x needs 2.5x |
3.0% | — | — | ecomadvertorialeverydayathletehypeus |
| Learning |
26W16_C250_V_2 · mashup · us £988
120245506609590365
Structurally sound; conversion gap vs UK likely LP/localisation
|
£988 | 2.18x needs 2.5x |
3.2% | 24% | 5% | mashupvslhealthfirstpodcastus |
| Learning |
26W18_C263_V_1 · aestheticLF · uk · brief LE
120246195294880365
Low CTR (1.6%) suggests brief type limits audience breadth; expertvoice hook intact
|
£480 | 2.10x needs 2.5x |
1.6% | 21% | 9% | aestheticLFugchealthfirstexpertvoiceuk |
| Learning |
26W18_USP_Slate_S_1 · ecom · us · workout
120246408751090365
Best Slate ad; 25bps below threshold; priceoffer hook is right but product unproven
|
£1,836 | 1.95x needs 2.2x |
2.9% | — | — | ecomproductshowcasetrendfollowerpriceofferus |
Top Losers
By spend at £400+ · all paused| Ad | Spend | ROAS | CTR | Tags | Diagnosis & Fix |
|---|---|---|---|---|---|
|
Loser
26W17_Meet_Slate_S_1 · ecom · UK · MOF · workout
120246066517430365
|
£483 | 0.77x thresh 2.5x |
2.2% | ecomproductshowcasehypeMOFuk |
Diagnosis
Funnel mismatch (MOF placement with cold-product creative) + hype hook failure (worst hook in account at 1.55x blended, W:0 L:2)
Fix
Replace hype hook with teaserhook. Move to TOF. Do not revive this ad — replace with a new hook variant.
|
|
Loser
26W19_C155_BundleLP_S_1 · ecom · US · workout advertorial
120246768330290365
|
£555 | 1.28x thresh 2.5x |
3.5% | ecomadvertorialhypeTOFus |
Diagnosis
LP mismatch (BundleLP destination with single-product creative intent) + hype hook failure. 3.5% CTR suggests clicks but no post-click conversion — LP is the primary suspect.
Fix
Replace BundleLP destination with product-specific LP. Swap hype hook to negativehook. Do not revive without structural changes.
|
|
Loser
26W11_USP_PolkaDot_S_3 · ecom · US · everyday cheeky
120243890820630365
|
£469 | 1.99x thresh 2.5x |
2.6% | ecomproductshowcasepriceofferTOFus |
Diagnosis
Marginal loser — 1 basis point below 2.0x threshold. Cheeky type may be narrowing the qualifying audience vs thong/highwaisted. PAUSED.
Fix
Test same concept with thong or highwaisted type before retiring. The priceoffer hook is correct — the issue is likely product type breadth.
|
26W19 Testing Sprint — Persona Analysis
46 ads · UK Testing + US Testing campaigns · £3,293 total spendPersona Scorecards
🥇 Trendfollower
3.75x
blended ROAS · £295 spend
4 ads · UK only · FreePair promo (CB: promo) · priceoffer hook · static only
⚠️ Promo-specific. ROAS inflated by Free Pair offer mechanics — not a like-for-like persona comparison.
🥈 Healthfirst
1.49x
blended ROAS · £1,368 spend
26 ads · UK + US · C250 podcast iterations + C111 + C182 · avg thumbstop 39.5% · avg p50 5.7%
✓ Strongest video engagement of all personas. C111 on correct LP hit 3.94x. BundleLP suppressing results.
🥉 Everydayathlete
1.38x
blended ROAS · £1,031 spend
10 ads · UK + US · hype hook throughout · workout product (Slate/LE) · all BundleLP
✗ Weakest non-promo persona. Hype hook + BundleLP + workout product = triple failure compound.
4th Minimalistcomfort
1.09x
blended ROAS · £341 spend
4 ads · UK + US · Organic156 hook iterations only · emphasizesolution + expertvoice hooks tested
– Hook swaps destroyed a proven winner (3.24x → 0x). Teaserhook is the mechanism — see below.
Critical Finding: BundleLP is the primary conversion blocker across all 26W19 tests
BundleLP destination = 1.46x blended ROAS across £1,935 spend. Standard LP = 2.06x blended ROAS across £1,350 spend. Thumbstop is virtually identical between both groups (37.3% vs 37.1%), confirming the video creative is equally strong — the LP is the conversion bottleneck. The clearest proof point is C111 (26W19_C111_BundleLP_V_1, aestheticDR/dramatisetheproblem/healthfirst): same exact creative, running simultaneously in UK Testing (BundleLP) and UK Scaling (standard LP). Testing/BundleLP = 1.56x. Scaling/standard LP = 3.94x. That is a 2.38x ROAS gap on identical creative. The C250 podcast iterations — which show 42–53% thumbstop across 16 video ads — are all pointing at BundleLP and averaging just 1.26x ROAS as a result.
BundleLP (most 26W19 tests)
Blended ROAS1.46x
Spend£1,935
Avg thumbstop (video)37.3%
Ads with any ROAS13 / 23
Standard LP (non-BundleLP)
Blended ROAS2.06x
Spend£1,350
Avg thumbstop (video)37.1%
Ads with any ROAS9 / 17
Proof: "26W19_C111_BundleLP_V_1 UK Testing" (120246569846440365) — £94.66 · 1.56x · BundleLP | "26W19_C111_BundleLP_V_1 UK Scaling" (120246702474490365) — £40.15 · 3.94x · Standard LP. Same creative.
Organic156 hook iterations confirm: teaserhook is the mechanism — replacing it destroys conversion
The original Organic156 teaserhook (A1/A7) runs at 3.24x UK and 2.26x US. W19 tested two hook replacements on the same creative: emphasizesolution (V1) and expertvoice (V2). Emphasizesolution: 0x ROAS in both markets with 15–17% thumbstop — lower than the original's 18–22%, meaning the hook change degraded attention-capture and killed conversion simultaneously. Expertvoice: 1.09x UK (best case), 0x US, 22–23% thumbstop — closer attention parity but still a near-total conversion collapse. This is a definitive finding: the teaserhook on Organic156 is not a stylistic choice, it is the conversion mechanism for the minimalistcomfort/resetdetox audience. Hook iterations on this creative should be approached by refining the teaserhook, not replacing it.
| Hook | Market | Spend | ROAS | Thumbstop | p50 | vs Original |
|---|---|---|---|---|---|---|
| teaserhook (original A1/A7) | UK | £423 | 3.24x | 18% | 5% | ● Baseline winner |
| teaserhook (original A7) | US | £642 | 2.26x | 22% | 5% | ● Learning zone |
| emphasizesolution (V1) | UK | £65 | 0.00x | 16% | 2.7% | ▼ −3.24x vs original |
| emphasizesolution (V1) | US | £112 | 0.00x | 17% | 2.1% | ▼ −2.26x vs original |
| expertvoice (V2) | UK | £100 | 1.09x | 23% | 2.7% | ▼ −2.15x vs original |
| expertvoice (V2) | US | £64 | 0.00x | 22% | 2.8% | ▼ −2.26x vs original |
Rule: Never swap the teaserhook on Organic156. If iterating, refine the teaserhook's opening line or visual reveal — do not replace the hook mechanic.
C250 podcast iterations show 42–53% thumbstop — best attention signal in the account — but BundleLP is blocking conversion
The 16 C250 podcast iteration videos (8 UK, 8 US) average 45.6% thumbstop UK / 45.5% thumbstop US — significantly above the already-strong original C250 (27% UK / 24–31% US) and the account video average (~26%). This means the iteration work has improved the hook. But average ROAS across all 16 iteration videos is only ~1.2x. The creative is working; the landing page is not. Only one non-BundleLP C250 iteration variant got meaningful spend (C250_It2_LP_V_3 UK: £118, 1.47x, 46% TS) — it's in the same suppressed ROAS range, suggesting standard LP still needs optimisation for these iterations. The highest p50 iteration in this cohort: 26W19_C250_It2_LP_V_1 UK (50.4% thumbstop, 6.1% p50) — needs budget on the correct product LP.
UK C250 Iterations (8 ads)
Avg Thumbstop45.6%
Avg p505.3%
Avg ROAS~1.26x
US C250 Iterations (8 ads)
Avg Thumbstop45.5%
Avg p504.9%
Avg ROAS~0.8x
Best iteration signal: "26W19_C250_It2_LP_V_1 UK" (120246572161540365) — 50.4% TS · 6.1% p50 · needs budget on correct product LP. "26W19_C250_It2_BundleLP_V_2 UK" (120246572216650365) — 45.5% TS · 5.8% p50 · 1.93x ROAS despite BundleLP.
Emerging Signals — Watch Next Sprint
C111 + dramatisetheproblem + healthfirst — 3.94x on correct LP at tiny spend
26W19_C111 (aestheticDR/vsl/dramatisetheproblem, healthfirst, UK, everyday) is running in both a testing adset (BundleLP: 1.56x at £94) and a scaling campaign adset (standard LP: 3.94x at £40). The LP comparison on identical creative is the clearest evidence in the dataset. C111 + dramatisetheproblem is an untested combination in the main account — if it holds above 2.5x at £400+, it becomes the account's first new winning hook angle in weeks. Test immediately: isolate C111 on standard LP, UK TOF, healthfirst, underwearskincare CB, everyday product.
Ads: "26W19_C111_BundleLP UK Scaling" (120246702474490365) — £40 · 3.94x · 18.6% TS · p50=9.0%
Ads: "26W19_C111_BundleLP UK Scaling" (120246702474490365) — £40 · 3.94x · 18.6% TS · p50=9.0%
C182 Georgia Symonds UGC — highest p50 in the entire 26W19 cohort with near-zero spend
26W19_C182_WL (ugc, georgia.symonds, dramatisetheproblem, healthfirst, everyday) has had almost no budget: UK at £21 (0x ROAS, 34.8% TS, 6.2% p50) and US at £7 (0x ROAS, 47.8% TS, p50=10.7% — the highest mid-body retention of any 26W19 creative). The 0x ROAS is due to insufficient impressions and BundleLP destination, not creative failure. C182 has the hook strength (48% TS US) and the retention (10.7% p50) to be a winner if given £400+ on a correct product LP. UGC + dramatisetheproblem + healthfirst mirrors the C111 angle — double down here.
Ads: "26W19_C182_WL US" (120246571217910365) — £7 · 47.8% TS · p50=10.7%
Ads: "26W19_C182_WL US" (120246571217910365) — £7 · 47.8% TS · p50=10.7%
FreePair promo statics — 3.46x and 4.47x but on cheeky type only
26W19_FreePair S1 and S2 (ecom + campaign, productshowcase, priceoffer, trendfollower, UK, everyday/cheeky, CB: promo) posted 3.46x and 4.47x respectively at £203 and £82 — both unclassified (under £400) but the highest ROAS signals in the week. The promo mechanic (Free Pair) is inflating ROAS above its sustainable baseline. However, the priceoffer + trendfollower + UK combination is confirmed working at conversion speed. Worth testing the same format against thong or highwaisted type (not cheeky) to see whether the conversion efficiency holds on higher-volume SKUs.
Ads: "26W19_FreePair_S_1 UK" (120246517857700365) — £203 · 3.46x | "26W19_FreePair_S_2 UK" (120246517894280365) — £82 · 4.47x
Ads: "26W19_FreePair_S_1 UK" (120246517857700365) — £203 · 3.46x | "26W19_FreePair_S_2 UK" (120246517894280365) — £82 · 4.47x
Everydayathlete persona needs a creative reset before further investment
The everydayathlete adsets ran C155 (hype/advertorial/workout/BundleLP) as the lead creative across both UK and US — a combination that is already confirmed as the account's weakest hook (hype: 1.55x blended) on its worst product (workout: 1.79x blended). With no video creative tested for this persona (all static), and all ads pointing at BundleLP, there is no valid signal here. Before investing further in the everydayathlete persona, test at least one video format (ugc or aestheticDR) with a teaserhook or negativehook on the standard LP. C168 (move.with.madeleine, negativehook, ugc) showed 24% TS but only 2.1% p50 and 0.53x ROAS — the body is collapsing post-hook.
Ads: "26W19_C168_WL uk" (120246570298870365) — £40 · 0.53x · 24% TS · p50=2.1%
Ads: "26W19_C168_WL uk" (120246570298870365) — £40 · 0.53x · 24% TS · p50=2.1%
26W19 Test Verdict Summary
| Persona | Spend | Blended ROAS | Avg TS (video) | LP Used | Verdict |
|---|---|---|---|---|---|
| trendfollower | £295 | 3.75x | — (static only) | standard | Promo-inflated. FreePair offer driving efficiency. Retest without promo mechanic. |
| healthfirst | £1,368 | 1.49x (suppressed) | 39.5% | mostly BundleLP | Strongest video engagement. C111 on correct LP = 3.94x. Retest all healthfirst on standard LP. |
| everydayathlete | £1,031 | 1.38x | 30.4% | all BundleLP | Hype hook + BundleLP + workout product = no valid signal. Creative reset required before retest. |
| minimalistcomfort | £341 | 1.09x | 19.6% | standard LP | Hook swap destroyed conversion. Do not iterate Organic156 hook. Teaserhook is non-negotiable. |
Note: 26W19 total spend includes £257 in untagged Quiz ads (Quiz1/Quiz2) excluded from persona analysis.