Research: Verifiable Software Economics

FW Delta Research

Publicly verifiable analyses of software economics, portability, automation and self-controlled infrastructure.

No paid rankings. No invented benchmarks. Methodology, sources and assumptions are published alongside.

8 flagship reports · open datasets · traceable methodology · versioned corrections

published reports: 8
numbered sources: 273
open datasets: 39
data cutoff: 10 August 2026

How this series came about

The first seven editions in this series were prepared over several months and published together on 29 July 2026. Data collection, modeling, source review and editorial review take time. FDR-2026-08 is the first regular monthly continuation of the series.

The month in an issue name marks its place in the series, not its release date. Both are stated separately on every report page, together with the data cutoff and the version, and the same applies to the machine-readable metadata. There are no backdated values.

A documentation or capability score measures only the object defined by the relevant report. It is not a general statement about compliance, security, quality or capture rate. Missing information means "not documented", not "does not exist".

Sorted by edition month, newest issue first. The series reads from the most recent question backwards; the real first publication date is on every card.

FDR-2026-08 Tracking Engineering Edition August 2026

Web Measurement Reliability & Recoverability Report 2026

FW Delta separates observed, corrected, modeled, invalid, and missing conversions and evaluates 30 measurement failure modes against ten control families.

First published: 13 August 2026
Data cutoff: 10 August 2026
Methodology: structured primary-source analysis with a failure-mode taxonomy, a 300-cell control matrix and six reference architectures
Objects examined: 30 failure modes; 10 controls; 300 control assessments; 6 reference architectures; 28 audit checks

Key findings

[OBSERVED] The GA4 Measurement Protocol is intended to supplement automatic collection rather than replace it; a fully server-side implementation can retain only partial reporting. [8]
[OBSERVED] A Measurement Protocol request can return HTTP 2xx even when its payload is malformed, incorrect, or not processed. [9][10]

View methodology and sources

Research standard and methodology

Statement classes, source hierarchy, score standard, handling of missing information, price and currency rules, rounding and conflict of interest.

Read the standard

Corrections and versions

Version policy, correction log, reporting route for factual errors and the current version status of every issue.

View the log

Research for technical decisions

New reports, benchmarks and technical analyses on SaaS economics, AI engineering and owned infrastructure.

Original research Public sources No sales mail

By subscribing you receive new analyses and updates from FW Delta by email. You can withdraw your consent at any time. Further information is available in the privacy policy.