v1.0.0 — Metricfixer Web GTM vs sGTM Benchmark

v1.0.0 — Initial Metricfixer Web GTM vs sGTM benchmark

This release contains the frozen dataset and reproducibility materials used by the Metricfixer publication Server-Side GTM Benchmarked: What It Really Improves—and What It Does Not.

Included

• 60 raw cold-load runs: 15 per architecture variant.
• Processed medians, P75, minimums, maximums, and comparisons.
• Portable Playwright runner for future reruns.
• Original runner retained for provenance.
• JSON Schemas, checksums, tests, Dockerfile, and CI workflows.
• Methodology, metrics, evidence matrix, limitations, data dictionary, and dated filter-list audit.

Main measured result

The proxy-only design retained the same browser JavaScript and event-request count as the web GTM-style design. The consolidated design reduced browser requests and JavaScript transfer while preserving four logical downstream deliveries in the local synthetic model.

Important limitation

The source package did not embed the exact runtime, browser, operating-system, or machine versions used on 31 July 2026. The frozen dataset is preserved as published; every future run records those values automatically.