Tracking & attribution engineeringClient work
A web + server GTM rebuild for headless Shopify: identity that survives the storefront-to-checkout hop, dedup, refunds and US opt-out consent.
Sounds familiar?
How I solve it
I build measurement that survives the storefront-to-checkout hop on a headless store: the Google client id and the ad click id are carried onto the Shopify order, one producer reports each purchase, a server-side leg sends the events from a stable source, refunds are fetched and subtracted, and US opt-out consent (Global Privacy Control and a reachable opt-out link) is wired without a cookie wall. On a headless build, that identity bridge is the whole job — without it a sale and the visit that produced it are two unrelated strangers.
The process
6 steps. Scope and a fixed price are agreed before the first one.
Export equals production, with no drift, before anything is touched.
The Google client id and the ad click id are written onto the order as cart attributes, and a checkout-side custom pixel reports with session context.
The old page-based purchase tags are retired and the duplicate producer is switched off, so two systems cannot report the same sale.
GA4 Measurement Protocol with a server-set first-party cookie, Meta CAPI and Bing, each with confirmed before/after evidence.
Successful refunds are fetched from Shopify on a schedule and sent to analytics, recorded so none can be sent twice.
A US opt-out package (GPC honoured before any tag loads, an opt-out link, no cookie wall) and a live report plus a daily monitor.
Handover
Built with
Proof
Client work Done for real clients. Client details are anonymised.
Before → after
In the 29 days before the rebuild, 612 sales reached analytics and only 39 carried any session source at all — 6%. After the identity bridge and the new server leg, traceability read 76% on the settled-day reading of 2026-08-19; duplicate purchase reporting went from +100% to 0%, and refunds were subtracted for the first time.
US headless-Shopify automotive-repair retailer
6% to 76%
sales traceable to source 6% → 76%; duplicate purchase reporting +100% → 0%; refunds subtracted for the first time (211 in 90 days); Google-attributed sales per day 0.1 → 10.5.
Read the case: 16 Days Before Shopify's Deadline: Rebuilding Measurement for a Headless StoreFixel Pixel
the server-side pipeline this rebuild can hand off to: one event mapping, a delivery receipt per event, and detectors for a dead connection, a dark store and a stalled delivery.
Super efficient and knowledgeable. He is one of the most thorough people I have ever hired in UpWork. His documentation is excellent.
What you can look at
A redrawn diagram of storefront → checkout → identity bridge → server leg → refunds, with a redacted excerpt of a before/after event table. The diagram and table are rebuilt on synthetic data for this page; the figures below are live reads.
Client details anonymised. Figures read from live platform APIs in August 2026. I am the founder of Fixel Pixel, the server-side pipeline named here as one option.
Price and timeline
A rebuild like this is scoped and quoted in writing as a Custom Engineering Project, from USD 1,500. If the cause or scope is not clear yet, the honest first step is a read-only Tracking Health Check (USD 395) or a Working Session (USD 195) on one question. A smaller, bounded piece can start at the Focused Fix minimum of USD 450. In the dated case the build was timed to beat a platform deadline: the new purchase leg went live on 2026-08-10, sixteen days before Shopify’s 2026-08-26 cutoff.
QuoteFixed before work starts
A note from Daniilbefore you decide
If your storefront and checkout already share one visitor identity and your refunds are already net, a rebuild adds work you don’t need — I’ll say so. And if the gap is only attribution modelling (view-through and different attribution windows), no capture fix will move it; a reconciliation will show you that instead.
— Daniil
Questions
Yes. The identity bridge runs through a checkout-side custom pixel and the Shopify order, so it does not depend on the theme or the native checkout scripts.
No. The storefront half is handed to your developer as a paste-ready patch package; the rest runs through supported paths.
Yes. Successful refunds are fetched from Shopify on a schedule and sent to analytics, checked against Google's validation service and recorded so none is sent twice.
Opt-out, not opt-in: GPC honoured before any tag loads and an opt-out link, with no cookie wall. This is technical implementation, not legal advice.
A dashboard is a view. This is identity and delivery: the same sale carries its source across the hop and reaches each platform once.
The first answer is free, within one working day. Or write directly: next@taskfordaniel.com
