Conversion data your ad platforms finally trust.
From clean browser-side tracking to full server-side builds: GA4, GTM, Meta CAPI with deduplication, consent mode v2, and every number reconciled against your store.
Definition
What is server-side tracking?
Server-side tracking is a measurement architecture where conversion events are sent to ad platforms from a server you control, instead of relying only on pixels firing in the visitor's browser. It matters because the browser side broke: Apple's iOS privacy changes, Safari's tracking prevention, and ad blockers now stop a large share of pixel requests from ever reaching Meta or Google, so the platforms optimise on incomplete data, CPAs climb, and ROAS looks worse than the business actually is. A correct server-side setup routes events through a server-side GTM container, sends them to Meta via the Conversions API and to Google as enhanced conversions, deduplicates browser and server events with a shared event_id so nothing is counted twice, respects consent mode v2, and is reconciled against the store's actual orders. Anlyto delivers this as a fixed-scope build across four tiers, from browser-side foundation to fully custom funnel tracking.
What a proper build ships with
Four tiers
Pick the depth your spend justifies
Fixed scope at every tier, and every tier is QA'd and reconciled. We will tell you honestly which one fits, sometimes the answer is the smallest one.
Browser-side Tracking
from $X
Clean, correct client-side tracking as your baseline. The right foundation if your container is a mess or you are starting fresh.
- GA4 rebuild + clean GTM web container
- Ecommerce dataLayer events, view_item to purchase
- Google Ads + GA4 conversions
- BigQuery export enabled
- Tracking plan + full QA
Best for: Brands that need a trustworthy client-side baseline first.
Server-side Tracking
from $X
The signal-recovery layer. Server-side GTM with CAPI deduplication, so the platforms see conversions the browser can no longer report.
- Server-side GTM on Stape or GCP
- Meta CAPI with event_id deduplication
- Consent mode v2 wired to your CMP
- Google Ads enhanced conversions
- Reconciliation against your store
Best for: Brands spending real money on Meta or Google that need recovered, trusted signal.
Full Tracking Setup
from $X
Browser plus server plus every channel you run, with durability and monitoring so the signal holds up as you scale spend.
- Everything in both tiers above
- TikTok, Pinterest, Snap CAPI as needed
- First-party signal durability
- Monitoring + alerting handoff
- 30 days post-launch QA
Best for: Brands scaling spend across multiple channels where every point of signal counts.
Custom Funnel Tracking
custom pricing
For funnels a template cannot describe: bespoke event design across every step, built and QA'd against your actual journey.
- Multi-step funnels + quizzes
- Subscription + recurring flows
- Headless + custom checkouts
- Cross-domain journeys
- Scoped against your funnel, quoted fixed
Best for: Stores whose funnel does not look like a standard product page to checkout path.
In every build
The parts template setups skip
Most tracking setups are an imported container nobody QA'd. The events fire, technically. Whether they fire once, correctly, with the right parameters, is the question that decides if your ad spend is efficient.
Ecommerce dataLayer
A documented dataLayer spec for every commerce event, implemented and validated so tags stop guessing what happened.
- view_item to purchase and refund
- Correct parameters + values
- Spec your developer can extend
Server-side GTM
First-party collection through a server container on Stape or GCP. Faster pages, fewer blocked requests, PII control at the edge.
- Stape or GCP hosting
- Meta + Google server tags
- First-party cookie handling
Consent mode v2
Wired to your CMP and tested in consented and non-consented states, so you keep modelled conversions without breaching GDPR.
- OneTrust, Cookiebot, Iubenda
- Both consent states QA'd
- EU + UK traffic covered
Attribution + UTM governance
A UTM convention your whole team follows, so channels stop leaking into direct and attribution reads clean across platforms.
- Naming convention + template
- Channel grouping rules
- Cross-domain parameter handling
BigQuery export
Raw event export switched on from day one. It is the foundation the reporting warehouse and profit dashboards build on later.
- GA4 to BigQuery linked
- Owned by your account
- Ready for the reporting build
QA + reconciliation
Every event verified firing once, correctly, then platform numbers reconciled against your store's orders before sign-off.
- Deduplication verified
- Consent-state testing
- Store reconciliation report
How a build ships
Plan. Build. QA and reconcile.
The tracking plan is approved before we touch anything, and nothing is handed off until the numbers reconcile against your store.
Plan
Current stack review, channels in play, and a written tracking plan covering every event, tag, trigger, and parameter, approved by you before any change goes in.
Build
Server container provisioned, web container cleaned, dataLayer implemented, CAPI and enhanced conversions wired with deduplication, consent mode v2 connected to your CMP.
QA & reconcile
Events verified firing once in consented and non-consented states, numbers reconciled against store orders, then publish, monitor, and hand off the documentation.
After launch
Tracking drifts the moment your site changes
A theme update, a new app, a checkout tweak, any of them can silently kill a tag, and you will not notice until the numbers stop making sense. Every tracking setup rolls into the Tracking Health retainer: ongoing monitoring, alerting, and fixes so the signal you paid to recover stays recovered.
FAQ
Tracking setup questions
What ecommerce founders ask us before a build, answered straight.
Stop feeding your ad platforms half the story
Book a 30-minute discovery call. We'll map your stack, tell you which tier actually fits, and quote it fixed. Not sure yet? Start with the tracking audit.
Continue reading
Related services
- Read more
Tracking Audit
Not sure what's broken? The audit inventories your tags, events, and consent signals and hands you a ranked fix list before you commit to a build.
- Read more
Reporting & Dashboards
The BigQuery export this build switches on feeds the reporting warehouse: profit per order, LTV, creative, and inventory dashboards.
- Read more
Retainers
Tracking Health keeps the build alive after launch: monitoring, alerts, and fixes as your store, theme, and app stack change.
Published
Last reviewed