Aug 16, 2026
Material connections, what the 2023 revision defined as clear and conspicuous, why a footer link is not enough, wording that a reader will actually parse, and why rel attributes satisfy Google rather than the FTC.
Shown, not told
Example scores real blogs against a fixed six-dimension rubric and publishes every number, including its own.
One send per teardown. The list is new and no teardowns have published yet, so you would be early.
Status · rubric v1.0
The method is published. The scoring has not started.
Six dimensions, thirty band definitions and a stated evidence standard are live and can be argued with today. No blog has been scored yet. When the first teardowns publish, every one will record the rubric version it was scored against.
Reference
Verified technical reference that teardowns cite instead of re-explaining. Every figure is sourced and dated.
Aug 16, 2026
301, 302, 307 and 308, which preserve the HTTP method, when a canonical or a 410 is the right tool instead, how to build a redirect map for a migration, and the one command that verifies it.
Aug 16, 2026
What actually differs between the three syndication formats, an element-by-element mapping, the autodiscovery markup that is the part actually scored, and why the rubric takes a position on full text.
Aug 15, 2026
SPF, DKIM and DMARC, what alignment actually is and why it is the part that breaks, the 0.3 percent spam threshold, one-click unsubscribe, and how to verify all of it from outside an account.
Aug 14, 2026
The three metrics, their exact good and poor thresholds, why they are assessed at the 75th percentile of real user data over a rolling 28-day window, and why a Lighthouse score is not the same thing.
Aug 9, 2026
Ghost wins on time to publish and default Craft. WordPress wins on the hub layer, the commercial surfaces and the pricing model. Compared dimension by dimension against the rubric.
Canon
Method
The rubric, v1.0
Six dimensions, what each is worth, and what a 1 and a 5 look like.
Evidence
What counts as evidence
Public evidence only. No analytics access, no interviews, nothing given in confidence.
Reference
Core Web Vitals thresholds
The exact LCP, INP and CLS bands Craft is scored against, and why lab data is not field data.
Audit
The paid version
Same rubric, your site, private, with the implementation sequence written out.
The list
Read the score before the subject does
Each send will be one scored blog, the three prioritized fixes, and what changed on the sites scored before it. Nothing else goes out.
Unsubscribe link in every send. The list is never rented or shared.