E-commerce Sales Funnel Audit

Export spend, revenue, and orders for Meta, Google, TikTok, and Amazon Ads from the last 90 days. If you don't run Triple Whale, pull from GA4 + each ad platform and reconcile against Shopify orders. Watch for iOS-attribution shortfall on Meta — pixel-only numbers under-count.

Blended MER = total revenue / total ad spend across all channels. Compare against your last quarter's MER and your contribution-margin breakeven. Anything below breakeven needs a channel-mix decision before next quarter's budget is locked.

Identify creative fatigue — frequency above 2.5 on Meta, declining CTR week-over-week, rising CPA. PMax asset-group reporting in Google Ads only shows top assets; pull the asset-group insights report for fuller signal. Queue 3-5 new creative concepts for the next sprint.

TACOS (total ad cost / total sales including organic) is the cleaner Amazon metric than ACoS — it captures the organic halo from sponsored placements. Pull from Helium 10 or Sellerise. Flag any ASIN with TACOS climbing while organic rank slips; that's a sign your ad spend is masking a listing problem.

For each channel: cut (below breakeven and not improving), scale (above target ROAS with headroom), or hold (within target band). Document the reasoning in your quarterly planning doc so the next audit has a baseline.

Healthy pop-up capture sits 3-6% of unique sessions for most DTC brands. If you're below 2%, the offer is weak or the trigger fires too early. Check mobile vs. desktop separately — mobile trigger timing is the most common miss.

Welcome flow should drive 3-6% of email revenue; abandoned-cart 5-10%. Pull the flow performance report in Klaviyo or Omnisend. Look for emails with open rates below 25% — those are subject-line rewrites, not full-flow rebuilds.

Subscribers who haven't opened or clicked in 90 days drag down deliverability for the rest of the list. Build the segment in Klaviyo (no opens, no clicks, no orders in 90d). If the segment is more than 10% of the active list, run a sunset flow before the next big send.

Three-email sequence: re-engagement offer, last-chance, then suppress non-responders. Suppress (don't delete) so you can still target with paid retargeting. Confirm the suppression segment is excluded from your campaign sends going forward.

Pull a sample of recent SMS opt-ins from Postscript or Attentive. Each record needs the express written consent timestamp, the disclosure language shown, and the IP address. Missing consent records are the #1 TCPA class-action exposure for e-commerce brands.

From a real phone (not Chrome devtools), add a SKU to cart, hit checkout, apply a discount code, complete with Shop Pay and a card. Watch for layout shift, slow shipping calculator, and discount-code field that's hidden behind an accordion. A theme update last week is the most common cause of a conversion-rate cliff.

Pull product analytics for the top 10 SKUs by sessions. Healthy add-to-cart rate is 8-12% for considered purchases, 15%+ for impulse. Outliers below 4% almost always have a hero-image, price-anchoring, or out-of-stock-variant issue.

Check LCP (under 2.5s), INP (under 200ms), and CLS (under 0.1) on the homepage, a top PDP, and the cart. Field data from CrUX is what Google uses for ranking, not lab data. Apps installed via Shopify App Store are the most common cause of INP regressions.

Score backlog items with ICE or PIE (impact, confidence, ease). Highest-leverage tests usually live on PDP or checkout, not the homepage. Confirm you have enough traffic to reach significance — under 50K monthly sessions, A/B tests below 10% effect size won't power.

Set the primary metric (revenue per visitor for revenue tests, conversion rate for funnel tests) before launch — changing it after seeing data is p-hacking. Plan for a minimum 2-week run to absorb day-of-week effects, longer if you're below 50K sessions.

90-day LTV is the realistic CAC ceiling for paid acquisition — 12-month LTV is too far out to plan against if you're cash-tight. Compare cohorts month-over-month; a declining trend means either acquisition quality slipped or repeat-purchase mechanics broke.

Healthy redemption rate is 15-30% of points earned. Below 10% means the program isn't earning loyalty, it's accruing a balance-sheet liability. Check redemption distribution by tier — if VIP tier never redeems, the rewards aren't compelling at the high end.

If you run subscriptions through Recharge, Smartrr, or similar, confirm the cancellation flow meets click-to-cancel parity — same number of clicks to cancel as to subscribe. California (SB 313) and FTC enforcement make this a real liability, not a cosmetic concern.

Smart-retry recovery should run 40-60% of failed charges. If you're below 30%, the retry schedule is too aggressive (cards lock) or your card-updater integration isn't pulling new card data. Involuntary churn from billing failure is recoverable revenue most brands leave on the table.

Export 1- and 2-star reviews from Yotpo, Okendo, or Amazon and tag by theme: sizing, quality, shipping, expectations-vs-reality. A theme that crosses 5% of total reviews on a SKU is a product-team escalation, not a CX response. Attach the export so the next audit has a comparison.