Skip to content
FW Delta Research

Research on software costs, measurement and owned infrastructure.

Publicly verifiable analyses with a disclosed methodology. Sources, assumptions and datasets are published alongside and there are no paid rankings.

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

Methodology Corrections and versions
published reports
9
numbered sources
328
open datasets
43
data cutoff
25 September 2026

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-2026-09 Tracking Engineering Edition September 2026

Measurement Platform Change Velocity Report 2026

FW Delta dates and codes 69 measurement-relevant platform changes from January 2025 to September 2026 and measures notice period, automation, opt-out and silent number changes per platform and reference architecture.

First published
26 September 2026
Data cutoff
25 September 2026
Methodology
structured primary-source analysis with a dated change register, notice-period calculation, five change types and an exposure model across six reference architectures
Objects examined
69 platform changes; 10 platforms; 65 computed notice periods; 6 reference architectures; 12 protocol items (69 objects in total)

Key findings

  • [CALCULATED] 69 measurement-relevant platform changes from ten platforms took effect between 1 January 2025 and 25 September 2026, 32 of them in 2025 and 37 in 2026 up to the cutoff.
  • [CALCULATED] The median notice period is 6 days across all 65 dated changes and 63 days across the 34 changes with advance notice.
View methodology and sources
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
Newsletter

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.