Server-Side Tracking & Marketing Data Architect

Overview We're a US DTC supplement brand on Shopify scaling paid acquisition, with a significant subscription/recurring-revenue business. We need an experienced server-side tracking and marketing-data specialist to own our measurement stack end to end, first to finish an in-progress remediation, then to architect our longer-term setup. This is a hands-on execution role, not advisory. You'll have full access and be expected to fix things, verify them, and report status without being chased. Our current stack • Shopify (Checkout Extensibility) • Meta Pixel + Conversions API • Elevar (server-side, our chosen hub) • GA4 + BigQuery • Triple Whale (attribution/analytics), Saras Pulse (BI) • Klaviyo (email), Postscript (SMS) • Stay AI (subscriptions / recurring orders) Phase 1 - Remediation (immediate) • Consolidate Meta tracking to one clean, deduplicated, consent-aware source via Elevar CAPI; eliminate duplicate source overcounting • Verify and close a consent-enforcement gap (no event data for users who declined consent), with proof via Meta Test Events • Raise Meta Event Match Quality on key events • Fix GA4 ecommerce funnel capture (mid-funnel events undercounting vs purchase) via server-side / webhook triggered events • Correctly handle subscription / recurring-order tracking (Stay AI) across Meta, GA4, and Klaviyo, including the decision of whether and how rebill orders should fire conversion events, and how first-order vs. rebill is distinguished • GA4 hygiene: channel grouping, custom definitions, Consent Mode (advanced) + behavioral modeling, key-events cleanup • Reconcile Shopify / GA4 / Meta / subscriptions against Shopify orders as the source of truth Phase 2 - Architecture & ongoing ownership • Design and run our server-side data layer as the single source of truth, with consent, suppression, and data hygiene enforced before any fan-out to destinations • Account for subscription/recurring revenue (Stay AI) in the source-of-truth model so reported numbers reconcile to actual Shopify orders • Evaluate and, if justified, implement sGTM (e.g., Stape), with a clear, documented cost/benefit recommendation (we will not build it by default) • Multi-channel server-side (Meta, GA4, Google Ads, TikTok, Klaviyo) with clean identity and deduplication • Ongoing measurement integrity, monitoring, and support as we scale spend You must have demonstrable, hands-on experience with • Shopify server-side tracking and the Checkout Extensibility / Web Pixels sandbox and its limits • Meta CAPI: deduplication (shared event_id), EMQ optimization, Events Manager diagnostics, dataset categories/restrictions • Elevar (configuration, server-side destinations, consent), required, not "willing to learn" • GA4 + BigQuery, Consent Mode v2 (advanced) and behavioral modeling • sGTM (server-side Google Tag Manager), ideally via Stape or self-hosted • Subscription/recurring-order tracking on Shopify (Stay AI, Recharge, or similar) and its impact on conversion data • Consent/privacy-aware tracking and why it matters (we treat consent compliance as non-negotiable) How we work / what we expect • You execute and verify, then report done / in-progress / blocked per item. We monitor; we don't want to manage keystrokes. • Responsiveness matters. We need someone who closes loops and communicates proactively, especially while we're pre-launch on ads. Slow or drifting communication won't work for this engagement. • Evidence over assertion: "fixed" means shown working (test events, reconciliation), not just configured. • We do not pursue tactics that evade platform classifications or terms; clean, compliant tracking only. To apply, please answer briefly (we filter on these) 1. A Shopify store fires more purchase events than begin checkout in GA4. What are the likely causes, and how do you fix it? 2. How do you ensure Meta CAPI doesn't send conversion data for users who declined consent, and how do you prove it's working? 3. How do you eliminate duplicate Meta conversions when pixel and CAPI both fire? 4. How would you handle subscription rebill orders in conversion tracking, should they fire purchase events to Meta? Why or why not? 5. When would you recommend sGTM/Stape over staying on a tool like Elevar, and when would you not? 6. Briefly: a server-side tracking project you owned end to end, the stack, what you fixed, and the measurable result. Generic proposals will be ignored. We're looking for specific, correct answers to the above.