Bare Bones · Acquisition-quality attribution

Which Meta creatives bring the most valuable new customers

Joining Meta ad creatives → real downstream Shopify value · Window: Apr 27 – Jun 26, 2026 (visible order window) · New-customer cohort

Headline findings

1 · Don't steer by Meta's numbers — it over-counts ~2×

Meta claims 4,717 purchases at 0.80 ROAS, but only 2,231 Shopify orders actually carry a Meta UTM and just 1,977 are genuinely new customers. Meta is crediting itself ~2.1× more conversions than carry its tag. True new-customer CAC is ~$141.92, not the ~$60 Meta implies. Only 7.9% of untracked orders even look social, so this is Meta inflation — not UTM leakage.

2 · UGC video acquires cheaper; Static images bring slightly better repeaters

UGC video brings customers ~16% cheaper ($147.75 vs $175.28 CAC) and is the volume engine. But Static-image creatives show higher early repeat (3.5% vs 2.5%) and similar 30-day LTV (~$48). The best ad on both axes is a Static "Weight Manager – Did You Know 3."

3 · Weight-Management & Motherhood hooks acquire the best customers

They post the highest early repeat (3.6% / 2.9%) at reasonable CAC. Gut-health / Longevity is the most expensive to acquire ($186.80 CAC) with the weakest repeat. Creator/TikTok ads acquire the cheapest customers ($94.26 CAC) but launched too recently to judge downstream value.

4 · The value signals are partly blind right now

Subscription starts are invisible at acquisition — only 3 of 3,458 new orders carry a native selling plan (Subscribe&Save runs through StayAI). And the 60-day wall truncates LTV. So "value" here = ≤30-day repeat + realized net revenue. Fixable: surface a {ad.id} macro + sellingPlan in checkout attributes.

Attribution coverage — how much can we actually see?

3,458
new customers acquired (window)
69.5%
carry UTM attributes (2,402); $113,502 / 70.6% of acq. revenue
88.2%
of Meta-tagged new custs joined to a specific creative (1,743 of 1,977)
50.4%
of ALL new-cust orders joined to a specific ad

Plain statement: 50.4% of new-customer orders and 88.2% of Meta-attributable orders are joined to a specific ad. Rankings below are based on that joined subset — ad-level Meta coverage (88.2%) is above the ~80% bar, so the creative/hook/format reads are reasonably trustworthy; whole-funnel coverage (50.4%) is lower because ~30% of orders are genuinely non-Meta.

The 1,056 untracked orders are mostly NOT Meta

Bucketed by Shopify's native first-visit source: Direct/unknown 609 (57.7%), Search/SEO 336 (31.8%), Email 15, and only 83 (7.9%) look paid/organic social. So the UTM gap is overwhelmingly genuine non-Meta traffic, with only a small (~83-order) sliver of possible Meta-UTM leakage to fix.

Meta reconciliation (window)

$280,583
Meta spend
4,717
Meta-reported purchases
2,231
Shopify orders carrying a Meta UTM
2.1×
Meta over-attribution (claimed ÷ UTM-tagged)

How the join & naming were mapped

Primary key failed. The leading numeric of utm_id (e.g. 120247118732340381_v2_s06) matched 0 of 201 live Meta ad IDs — it's a hard-coded ID carried over when ads are duplicated. Per your fallback, the join dropped to ad-name level: utm_term → Meta ad name (exact-normalized + unique-prefix). Confidence is high because both use the identical naming convention; 88.2% of Meta orders mapped.

utm_campaign — [NM] IBB - Prospecting - Scaling - CBO - LC Rebuild → [NM] = agency (NordMedia)IBB / CDM = product lineProspecting / Retargeting = funnelScaling / Testing = strategyCBO / ABO = budgetLC = objective/landing

utm_term — IBB - The Weight Manager Manager - Did You Know 3 - Nord Static - Paid media Lander - Brand Page → seg1 = productseg2 = persona/hook or creatorseg3 = concept/variantseg4 = vendor + FORMATseg5–6 = landing + page type

Format from seg4: "Static" → Static image · "UGC"/"TikTok" → UGC video · "Nord Video" → Studio video. Hook theme clustered from persona/concept keywords (Motherhood, Weight management, Gut health/Longevity, Collagen/Beauty, Creator/Influencer, Price/Promo). Funnel: essentially all spend is Prospecting — no meaningful retargeting campaign ran in-window. Correct any misread tags and I'll re-cluster.

Ranking by FORMAT

FormatNew custsSpendCAC (true)30-day repeat30-day LTVLTV:CAC
UGC video1,169$172,714$147.752.5% (n=317)$47.990.32
Static image555$97,281$175.283.5% (n=344)$47.300.27
Studio video19$9,189$483.61———

CAC = window spend in that format ÷ mapped new customers (true CAC, new-customers only). 30-day repeat & LTV measured on customers acquired by May 27 (n shown). Studio video sample is tiny/recent.

Ranking by HOOK theme

FormatNew custsSpendCAC (true)30-day repeat30-day LTVLTV:CAC
Motherhood698$107,770$154.402.9% (n=242)$47.660.31
Gut health / Longevity446$83,313$186.802.4% (n=170)$46.100.25
Weight management331$54,200$163.753.6% (n=250)$48.460.30
Creator / Influencer233$21,963$94.26———
Price / Promo35$9,223$263.53———

Creator/Influencer (TikTok) and Price/Promo launched mostly mid-June → too recent for 30-day value (shown as —), but Creator has by far the lowest acquisition CAC.

Ranking by INDIVIDUAL CREATIVE

Proven set only — creatives with ≥25 acquired new customers AND ≥15 with 30-day value observability. Sorted by early repeat, then 30-day LTV.

Creative (ad name)NewSpendMeta CPACAC (true)30-day repeat30-day LTVLTV:CAC
ibb - leah x - the mother - bb ugc - back to work - video - paid media landing page - partner page30$6,056$41.76$201.865.6% (n=18)$68.680.34
cdm - the health manager - 099 collagen without the hassle 3 - nord static - paid media lander - brand page135$18,974$60.62$140.554.1% (n=74)$44.110.31
ibb - the weight manager manager - did you know 3 - nord static - paid media lander - brand page241$33,745$60.26$140.023.6% (n=223)$48.510.35
ibb - the mother - daily routine mom day 2 - bb ugc video - paid media lander - brand page529$72,947$69.34$137.903.2% (n=158)$46.390.34
ibb - living beyond measure - march - busy mom - the mother - bb ugc - video - paid media landing page - partner page79$13,500$42.86$170.892.7% (n=37)$43.410.25
ibb - the health manager - body needs 4 - nord video bb ugc - paid media lander - brand page180$27,908$65.36$155.041.6% (n=63)$47.260.30

Scaled but too recent to judge value (≥25 new, <15 observable)

Creative (ad name)NewSpendMeta CPACAC (true)30-day repeat30-day LTVLTV:CAC
ibb - tiktok kelsilee06 - 7648325093973658894 - paid media lander - brand page141$11,198$39.85$79.42———
ibb - the health manager - 100 fits well 3b - nord static - paid media lander - brand page66$15,591$85.20$236.23———
ibb - tiktok shopwithshawn_ - 7648813718972779807 - paid media lander - brand page60$5,633$40.82$93.89———
ibb - the health manager - 090 longevity nourishers 3 - nord video bb ugc - paid media lander - brand page52$10,615$72.70$204.130.0% (n=10)$55.080.27

Promising but unproven (5–24 new customers)

IBB - Madeline Bishop - The Mother - BB UGC - — 24 new, CAC $274.72IBB - TikTok laceylikestotravel - 764684988931 — 22 new, CAC $141.58IBB - The Weight Manager Manager - Did You Kno — 18 new, CAC $205.51CDM - 129- v1 - summer sale - - Nord static — 16 new, CAC $201.00CDM - The Health Manager - 116 flavor first 3 — 15 new, CAC $210.42CDM - The Health Manager - 099 Collagen withou — 14 new, CAC $187.07IBB - 127 - v2 - summer sale - - Nord static — 14 new, CAC $277.67IBB - Living Beyond Measure 1 - The Mother - B — 10 new, CAC $101.64IBB - The Health Manager - 104 Reminder Style — 10 new, CAC $237.28IBB - ariel d - The Mother - BB UGC - post par — 9 new, CAC $140.61IBB - The Health Manager 121 Four Quadrants 3 — 7 new, CAC $268.45IBB - The Mother - Daily Routine Mom Day 1 - — 5 new, CAC $167.24IBB - The Mother -117 Protein Hack - Nord BB — 5 new, CAC $457.90CDM - 126_V2_BareBones_CDM-Health Manager-TheG — 5 new, CAC $108.90

Biggest disagreements: Meta's verdict vs realized Shopify value

Across the board Meta's CPA ($42–69) runs 2–3× cheaper than true new-customer CAC ($138–202) — Meta credits returning buyers and view-throughs it can't actually claim. Within that, the relative ranking still diverges:

Hidden gems — Meta underrates, Shopify shows value

Meta gives them an average-to-poor CPA, but they bring the best repeat buyers.

  • CDM – Collagen “099 without the hassle” (Static)Meta CPA $61 (middling) but the highest early repeat of any scaled ad, 4.1%. Collagen acquirer worth leaning into.
  • IBB – The Mother “Daily Routine Mom Day 2” (UGC video)Meta's most expensive scaled ad by its own CPA ($69), yet it's the volume workhorse (529 new) with solid 3.2% repeat and the lowest true CAC ($138).

Overrated by Meta — cheap clicks, weak buyers

Meta loves them (low CPA) but the customers don't come back.

  • IBB – Living Beyond Measure “Busy Mom” (UGC video)Meta CPA just $43 (looks like a star) but lowest repeat & LTV of the scaled set (2.7% / $43).
  • IBB – Health Manager “Body Needs 4” (UGC video)Meta CPA $65, but downstream repeat is only 1.6% — acquires triers who don't re-order.

Both agree — genuine winners

Strong on Meta's CPA and on real Shopify value.

  • IBB – Weight Manager “Did You Know 3” (Static)Meta CPA $60, true CAC $140, 3.6% repeat, best LTV:CAC of the big ads (0.35).
  • IBB – leah x “The Mother / back to work” (UGC video)Smaller (30 new) but Meta CPA $42 + top repeat 5.6% and top LTV30 $69.

Caveats & assumptions

  1. 60-day data wall. Shopify exposes only ~60 days of orders here (no read_all_orders scope), so the literal "first order in last 6 months / 90-day LTV" isn't possible. Window used: Apr 27 – Jun 26, 2026.
  2. Subscription starts unmeasurable at acquisition. Only 3 of 3,458 new-customer first orders carry a native sellingPlan; Subscribe&Save is run through StayAI, so the highest-confidence LTV signal you wanted is not visible at the point of acquisition. Weighting therefore shifted to repeat + realized LTV.
  3. LTV is realized 30-day, net of discounts & refunds (uses netPaymentSet), measured only on customers acquired by May 27 so every cohort has ≥30 days. Recent cohorts (TikTok, Summer Sale) show "—" for value.
  4. Ad-ID join failed (0%) → name-level join used (88.2% of Meta orders). ~234 Meta orders had UTMs too truncated (e.g. just "IBB") to map to one creative.
  5. True CAC slightly overstated at format/hook/ad level: spend is full, but the new-customer denominator only counts mapped customers (~88.2% of Meta). Real CAC is modestly lower.
  6. Meta CPA vs Shopify CAC aren't apples-to-apples (Meta = all purchases incl. returning + view-through; Shopify = new customers only). The ~2× gap is the finding, not a defect.
  7. Early-repeat rates look low partly because a broth/collagen repurchase cycle (~30–60 days) exceeds the observable window for June cohorts.

Generated from Shopify Admin order custom attributes + Meta Ads (acct 196212110934535). New customer = first order in window (customer first seen Apr 27+). All dollar figures USD.