If you're responsible for implementing tracking solutions on a publisher's website/app or for feeding the publisher's data warehouse with reliable data, you've seen it: Safari ITP and ad blockers mess with identifiers and tracking requests, consent states cause signal loss on top of that – and all of it leads to KPI drift. Meanwhile, tags and SDKs quietly trigger performance regressions or break with the next release.
In this technical deep dive, we'll walk through real-world publisher setups, typical failure modes, and proven patterns for building tracking and data pipelines that are stable, observable, and compliant – across web and apps, and aligned with the relevant regulatory environment.
What you'll take away:
- Publisher reference architectures: client-side vs server-side vs hybrid – when each approach makes sense
- Tag bloat reduction and Core Web Vitals protection without constant firefighting
- A debugging framework for "Why don't the numbers match again?" (drops, duplicates, cross-domain, SPA, app↔web journeys, etc.)
- Clean consent state modeling (opt-in / opt-out / undecided) for fewer errors and stronger auditability
- Identity & persistence across brands and domains – incl. common Safari/ITP pitfalls
- Schema drift prevention via tracking plans and data contracts, so reporting is actually reliable
- Decision logic by use case & regulation: EU (GDPR/ePrivacy) vs US vs global/mixed
Who this is for: Platform/Engineering, SRE, CMS/Integration Engineering, Data Engineering, Analytics Engineering, BI, and Experimentation/Measurement teams in publishing.
Speakers
.webp)
