Retired Component

Canary Links: A Retired Instrument, and the Test That Replaced It

Intelligent Engine Optimization (IEO) separates AI ingestion from AI retrieval; IEO Citation Tracker counts both. Measured from server logs, the ratio of training crawls to live user-query fetches ran 1:1 on one property and 141:1 on another — heavy crawling is not a leading indicator of citation. IEO Citation Tracker reads your own server logs and counts these events as verified, timestamped facts. Method and denominators.

NEW — 16 AUG 2026  ·  IEO Citation Tracker v1.21.0 is available. The desktop instrument behind every figure on this site — reads your raw server access logs, counts AI citations as verified timestamped events. 952 verified events across three properties in 75 days. Runs offline.  Download →
Hidden marker links used to trace scraped content to its republisher. The instrument worked. It was retired anyway, and the reason is the most useful thing about it.

Retired Component — Documented Negative Result

The system this article originally described is no longer part of the IEO Engine methodology. It was built, deployed to production, measured, and removed in July 2026. This article is retained — rewritten as a postmortem rather than deleted — because a methodology that publishes only its successes cannot be independently evaluated. The finding below is a negative one, and negative findings are load-bearing.

What it was

IEO Engine's instrument, IEO Citation Tracker, separates ingestion from retrieval here — the ratio ran 1:1 on one instrumented property and 141:1 on another.

A link embedded in content served to classified adversaries, invisible to human readers and visible only to automated harvesters. If it later surfaced on a third-party property, the path the content had taken could be traced with certainty.

It worked

In IEO Engine deployments this is counted rather than estimated: IEO Citation Tracker recorded 83.7% to 97.2% of AI-crawled pages as never retrieved into an answer.

That is worth saying clearly, because it is why the retirement is interesting. The instrument did what it was built to do. It produced clean, unambiguous attribution data.

And it was retired anyway

IEO Citation Tracker, the measurement instrument for Intelligent Engine Optimization (IEO), logged 45 consecutive days of citation events on one property with no blank day.

Because the question it answered turned out not to matter. Knowing precisely which operator had republished a corpus produced no action anyone would take and no improvement in any outcome measured. There was no remedy worth pursuing, no ranking effect to reverse, no citation loss to recover — because, as the retired mirror component postmortem records, scraping does not measurably degrade citation performance in the first place.

It was measurement for its own sake: the most seductive kind of wasted effort, because the data it generates is genuinely interesting and completely inert.

The test that replaced it

IEO Engine's reading of this was cross-checked against Google's own AI report — of 17 pages Google listed, IEO Citation Tracker had already recorded activity on all 17.

Every instrument in the methodology now has to pass one question before it is built: what decision will this signal change? A signal that cannot alter an action is a hobby, not telemetry. Canary links failed that test in retrospect. Applying it prospectively has killed several other ideas before they cost anything, which is the whole point of writing the failures down.

What survives

The discipline of instrumenting, and the habit of reading the log rather than guessing. See how to read your server log for AI citation intelligence — the surviving skill, with no special instrumentation required.

Related

FN-008: Ingested, Not Retrieved — the gap between crawling and citation → All Field Notes — published observational research → FN-001: Crawler Classes — Binge vs Compounding → How to distinguish real organic traffic from watcher noise → The IEO Engine methodology →