← Work · Case 03
Search monitoring that diagnoses before it alerts
n8n collects, an AI step diagnoses, a human decides.
StatusProduction
Typen8n workflows
RoleConcept, design and build, with AI agents
01 · Problem
Problem
Search Console data across several properties is too much to read by hand and too noisy to trust blindly. Raw alerts either get ignored or trigger panic changes.
02 · Approach
Approach
Scheduled n8n workflows pull the data on daily, weekly and monthly runs. Each signal gets an AI diagnosis with its evidence attached, lands as a task in Notion and waits there. A person decides what, if anything, changes.
03 · Architecture
Architecture
04 · Guarantees
Guarantees
- 01No workflow changes a live site.
- 02Every task carries the data it was based on.
- 03Unverified diagnoses are labelled as unverified.
- 04One run per schedule: no duplicate executions.
05 · Stack & decisions
Stack & decisions
| n8n | Visual workflows that are easy to audit and to roll back. |
|---|---|
| Google Search Console | The source of truth for search performance. |
| Google Sheets | A history anyone can open and check. |
| Notion | Tasks with a status, an owner and the evidence attached. |
| Claude | The diagnosis step, and only that step. |
06 · Proof
Proof
- principledetect → compare → verify → human decision
- the system never changes a live site.
- the human decides; the agent diagnoses.
- hard rules, each one paid for
- 1no bursts of manual runs: one recovery run, only if today has none
- 2never re-run weekly or monthly jobs blind: appends would duplicate
- 3deploy by procedure: read live → versioned backup → minimal change
- → local simulation → write → check status → read back → drift check
- 4data from a crashed run is unreliable
- 5fact ≠ hypothesis ≠ verified root cause ≠ unknown
- 6output is a file and a task. never a live change.
Excerpt from the real system, translated and sanitized: no names, identifiers or data.