Aether Instruments
Digital products platform · Data unification + checkout CRO
Aether Instruments sells digital products globally via a WooCommerce store, running paid acquisition through Google Ads across multiple European and international markets. The product catalog spans several categories with different price points and purchase cycles.
At audit, three separate reporting systems — Google Ads, WooCommerce, and GA4 — each reported different revenue for the same period with no reconciliation mechanism. No checkout funnel step tracking existed. A/B tests had been attempted on the product pages, but with no reliable attribution layer, none could be validated. The task: build a unified data foundation before any growth work could begin.
Three systems, three revenue figures, zero reconciliation. Each system attributed conversions differently — creating the illusion of data while making decision-making structurally impossible.
Attributed revenue included view-through and cross-device overlap — inflated vs. actual orders
Ground truth for orders, but no marketing source data — couldn't identify which campaigns drove which sales
Lost session attribution due to ITP, redirect chains, and inconsistent UTM tagging across ad placements
- →Custom GA4 data layerWooCommerce events instrumented
Purchase, add-to-cart, checkout step, and payment events sent to GA4 with consistent parameters and order IDs
- →WooCommerce → Google Ads linkRevenue attribution by campaign
Order value and product data passed from WooCommerce to Google Ads via GA4 export — ROAS calculations now match actual revenue
- →Cross-channel dashboardUnified weekly reporting
Single automated dashboard pulling spend from Ads, orders from WooCommerce, and sessions from GA4 — one revenue figure, daily refresh
- →UTM standardizationConsistent source tracking
All campaigns, ad groups, and creatives tagged with a standardized UTM schema — eliminating the misattribution from inconsistent parameter naming
With step-level funnel tracking in place, the 1.5% store conversion rate could be diagnosed properly. The largest drop-off: 41% of users who entered checkout exited on the payment step. Six A/B tests were run on the checkout flow, prioritized by drop-off volume.
Three systems reporting different revenue figures for the same period. Google Ads claimed ROAS of 4.1×. WooCommerce reported lower revenue. GA4 showed a third number. No one knew which was correct — so no one could act on any of them. The problem was not performance; it was the absence of a shared ground truth.
A 1.5% store conversion rate is a symptom, not a diagnosis. Without funnel step tracking — add to cart, checkout initiation, payment entry, order confirmation — there was no way to locate the drop-off. Once each step was instrumented, the pattern was clear: 41% of users who initiated checkout exited on the payment step.
Linking WooCommerce order data directly to Google Ads campaigns let the bidding algorithm optimize toward actual revenue, not proxy conversion events. Campaigns that looked efficient on click-through metrics turned out to have low average order values. The revenue link rearranged the priority of every campaign in the account.
Running A/B tests on checkout variants while attribution is broken produces results that can't be trusted. Every test variant inherits the same measurement error. Fixing data architecture first — before running tests — meant the checkout CRO results were actually actionable, not just statistically suggestive.
Three disconnected systems reconciled into a single source of truth. Store conversion rate improved from 1.5% to 2.8% (+87%) across six A/B test iterations on the checkout funnel. Revenue attribution aligned across Google Ads, WooCommerce, and GA4 for the first time — making ROAS calculations and campaign prioritization trustworthy.
CRO: 6 A/B tests · largest win — payment step redesign (+18% recovery)
Conversion rate: 1.5% → 2.8% · revenue attribution: 100% cross-system match
Client name is a pseudonym. Report ID is internal. Metrics reflect specific account conditions — results vary by market, budget, and funnel maturity.