SEO Field Notes
Real SEO work, shown one finding at a time
Every entry is a single thing we found in the data on a live website, what we did about it, and the metric we are watching to see if it worked. No changelog padding.
Audio Summary
Listen to Field Notes
Kodi gives a two-minute tour of how each note starts with one measured finding, shows the evidence, and keeps the next result open until the data answers it.
Read by Kodi, AI co-architect
Read transcript
SEO Field Notes is where Real SEO Life shows the work one finding at a time.
These are not broad case studies or padded work logs.
Each Field Note starts with one specific thing found in real Search Console or warehouse data.
Then it shows the evidence.
The actual numbers, tables, or charts that produce the finding are included right in the note.
And every note ends with an open loop.
What was changed?
What metric is being watched?
And when should we know whether it worked?
The current notes show both wins and mistakes.
One redirect audit fixed seven URLs, but missed the one Google was actually ranking.
One AI crawler generated more than 43,000 requests while barely reading any meaningful content.
On the same restaurant network, a machine-readable data layer was published, and AI systems came back and consumed the entire layer multiple times.
Another experiment showed why crawler identity has to be verified before a traffic spike is treated as real interest.
There is also an example of an unattended AI routine publishing directly to a live site where the most important lesson was not what the AI produced but what its safeguards prevented it from doing.
The AS400System network documents a major Google visibility drop following the August update.
Instead of immediately rolling everything back, the experiment is being watched to see what survives and what recovers.
And Dataset SEO records the opposite kind of lesson.
A page that was approaching page one was re-optimized for broader search intent and dropped 65 positions in two weeks.
That is really the purpose of Field Notes.
Real websites.
Real data.
Real decisions.
The wins get recorded, but so do the misses.
Instead of pretending SEO is a clean sequence of best practices, Field Notes documents what actually happened, what was changed because of it, and what the data says next.
That makes it less of an SEO blog and more of a working laboratory notebook for search, AI discovery, and website systems.
Every note, with its source data table as structured rows:
/ai/field-notes.json
open feed
How to read a field note
Finding first. Data shown. Open loop at the end.
One real thing
Each note covers a single observation from live Search Console or warehouse data, not a summary of a whole work session.
The numbers, in the note
Every field note carries the actual table or chart the finding came from. No data payload, not a field note.
What we are watching
Each note ends with the specific metric that will show whether the change worked, and roughly when to check it.
All Field Notes
8 TotalThe Map Everyone Builds On Is Not Google's
Google's place data needs a key and cannot be kept. Open map data downloads whole and can be. That one rule decides which map almost every tool outside Google is built on.
The Redirect Audit That Checked the Wrong List
A cleanup session added seven redirects and missed the one URL Google had actually indexed. The site's best-ranking page went from 158 impressions at position 31 to nothing, and th…
An AI Crawler Read 39,000 Pages of Nothing on Our Site
A verified OpenAI crawler pulled 43,309 requests from one restaurant network in a single evening. Almost none of it was content.
We Published a Dataset for Machines. They Read It Five Times.
A JSON record for every venue, a category taxonomy, a federation manifest and an llm.txt. In 25 days AI systems read the whole layer five times.
What a Crawler Claims Versus What It Can Prove
Before treating a traffic spike as interest, confirm the visitor is who it says it is. Two of ours were lying, and three honest ones could not prove anything.
An Unattended Routine Published to a Live Site. The Receipt Is What It Refused to Do.
A weekly research routine ran with no editorial review step in front of it and wrote three files to production after publish permission was granted. The useful record is not the ar…
Google Cut the AS400System Family 86 Percent. We're Not Rolling It Back.
Every established AS400System.com property fell together the day after Google's August spam update finished, while indexing and crawler traffic stayed alive.
We Re-Optimized a Page for Everyone's Intent and Tanked Its Rankings
A young site was two weeks from page one for its core term. A reasonable-sounding optimization pass for mainstream search intent dropped it 65 positions in fourteen days.