Weekly Funnel Audit

Week of Jul 6 – Jul 12

Every step of the acquisition funnel, 13 weeks deep, with this week's CAC change traced to the step that caused it. Generated Jul 14, 8:45 AM PT.

CAC fell $1.44 to $107.27 only because eCPM dropped 9.4% — a media-cost tailwind masked a homepage-led CVR erosion that would have added ~$9.

CAC fell $1.44 to $107.27 as a $10.65 eCPM tailwind was largely offset by a $6.20 NC CVR headwind and a $3.02 CTR headwind — a media-cost win masking site-conversion erosion.

Callouts

🔴 Homepage CVR — fix this week. Home collapsed 2.947% → 2.349% (-20.3%) on 22,481 sessions, the single largest LP move and the biggest contributor to the $6.20 CVR headwind. Audit home hero + nav for a Jul 6 change; if none, run a hero/CTA test by Friday.
🔴 Blended CAC still $22 above target — do not celebrate the -$1.44. Pulse blended $107.27 vs $85 target; tracker D2C CAC $105.17 vs $90; blended incl. Amazon $79.36 vs $75. eCPM relief is the only working lever — bank it, don't spend against it.
🟡 Listicle LP -13.4% — starve, don't kill. Listicle CVR 1.564% → 1.355% on 28K sessions while Rise-2 held (1.203% → 1.255%). Cut listicle paid traffic allocation 25% toward Rise-2 until CVR recovers above 1.5%.
🟡 Product page ripping at 3.821% (13-week high) — route more warm traffic to it. Home is broken, PDP is peaking on 34,340 sessions. Shift homepage nav + email links toward PDP entry while home is under repair.
⚪ Watch the Jul 6 Meta spend-definition change. eCPM -9.4% and CTR -2.8% coincide with the name-based exclusion going live with a Mar–Jul backfill. Confirm the eCPM win is auction-real, not a mix artifact, before extrapolating.
Checkout / ATC / co2p panels degraded since May 18 (GA4 begin_checkout) — directional only, no orders issued off them. Bounce panel step-change is instrumentation, not behavior.

Watch next week

Homepage CVR Mon–Wed: needs to reclaim 2.5%+ on comparable sessions or the -20.3% is the new regime, not a blip.
eCPM: does it hold in the $40–45 band or drift back toward $45+? The entire -$1.44 CAC move depends on this.
NC CVR exit trajectory: last week 1.605% was the top of the post-May-10 band (1.29–1.61%); if next week prints below 1.45%, we've stepped down again inside the broken regime.

Decomposition

Why CAC moved this week

Exact arithmetic, not attribution: blended CAC is the product of these four steps, so their contributions sum to the week's CAC change to the cent. Method in the appendix below.

Step by step · 13 weeks

The acquisition chain

Step by step · 13 weeks

On-site: where sessions become customers

Cross-checks

Measurement health

Agent analysis

The story of the week

This is an offsetting week, not a flat one. The waterfall shows media cost (eCPM) pulled CAC down by $10.65 as eCPM fell from $44.76 to $40.56 (-9.4%), returning to the middle of the $40–55 target band. That single step is the entire good news. Traffic leverage was exactly flat at 2.247x and contributed essentially $0.

Working against it: NC CVR dropped from 1.605% to 1.515% (-5.6%), adding $6.20 to CAC, and CTR slipped from 1.142% to 1.110% (-2.8%), adding another $3.02. Net: -$1.44. Without the eCPM relief, CAC would have risen ~$9 this week.

Tracing the CVR step: the landing-page split points at the homepage. Home CVR collapsed from 2.947% to 2.349% (-20.3%) on 22,481 sessions — the largest LP move of the week. Listicle also fell 13.4% (1.564% → 1.355%). Rise-2 was flat-to-up (1.203% → 1.255%) and product-page CVR actually hit a 13-week high at 3.821%. So the CVR drag is concentrated in the warm-pool entry point (home) plus one paid variant (listicle), not a broad-based site failure.

Regime context: NC CVR at 1.515% is back below the 2% floor and squarely inside the post-May-10 low-CVR regime (1.29–1.61% for the last seven weeks). Last week's 1.605% reading was the top of that band, not a new regime — this week is mean-reversion within the broken regime, not a fresh break. CAC at $107.27 remains $22 above the $85 pulse target and the story of the quarter is unchanged: cheaper media is the only lever currently working.

Confounders checked

Traffic mix: Paid share of sessions barely moved (46.2% → 46.8% per the traffic-quality lens; 46.3% → 46.3% in the chain series). Mix cannot explain the 5.6% blended NC CVR drop — within-bucket CVR fell too (paid 1.33% → 1.19%, warm 2.70% → 2.33%). The CVR softness is real, not a composition artifact.
Measurement — bounce panel: Paid bounce 10.67% and warm 15.29% is the second consecutive week of a step-down from the ~24%/30% baseline. This is almost certainly an instrumentation change, not behavior, and should not be read as engagement improving alongside the CVR drop.
Measurement — checkout funnel: ATC (-6.6%), session→checkout (-5.8%), and checkout→purchase (-1.2%) all softened, but GA4 begin_checkout has been degraded since May 18. These panels are directional only and cannot corroborate or refute the NC CVR move; NC CVR itself is sourced from Shopify D2C new customers ÷ GA4 sessions and remains the acquisition-truth number.
Spend-definition change (Jul 6): The Meta name-based exclusion list went live this week with a Mar–Jul backfill. Pre/post levels are described as consistent, but the CTR -2.8% and eCPM -9.4% moves are large enough that some portion could be a mix effect of which Meta campaigns are now counted — consistent with, but not proof of, a real auction-cost decline.

Regime changes

Events ledger

Numbered flags appear on every chart above. To add an event, edit reports/funnel-audit-events.json in the marketing-pulse repo — anything that could move a funnel step belongs here (site changes, tracking changes, promos, data outages).

Method appendix — step definitions & decomposition

The identity: CAC = spend/NC = (spend/impressions) × (impressions/clicks) × (clicks/sessions) × (sessions/NC) — so a change in CAC is exactly the sum of the four steps' log-changes, converted to dollars with the log-mean. Contributions reconcile to the cent by construction; the generator refuses to publish if they don't.
Spend & New Customers: Daily Spend Tracker — the canonical blended-CAC source (spend ÷ Shopify D2C new customers). Never a sum of Northbeam channel CACs.
eCPM & CTR: Northbeam paid media (Clicks Only model, cash accounting), Amazon excluded. eCPM uses tracker spend over NB impressions, so it's an effective blended rate, not a platform-reported CPM.
Traffic leverage: GA4 sessions ÷ NB paid clicks. Rises when organic/warm traffic carries more of the load per paid click; falls when the site depends more on paid.
NC CVR: Shopify D2C new customers ÷ GA4 sessions. Never Omni Total Orders (includes renewals).
Paid vs warm: Paid = rise-2 + listicle landers (cold prospecting); warm = everything else. Different intent pools — each is compared to its own history, never to each other.
Checkout steps: GA4 begin_checkout degraded since May 18, 2026 — those two panels are directional only until tracking is repaired.