Please purchase this course, or sign in if you’re already enrolled, to access the course content.
Lesson overview
This practical lesson brings Lessons 2.2 and 2.3 together into one written procedure. You go from a single pasted page to your whole corpus in the Entity Extractor, export the two files the rest of the course needs, and finish with your own Clarity score.
Each step pairs a screenshot of the real tool with what you should see on the demo site, so you can tell straight away whether your run is behaving: a one-document estimate of a handful of units, an entity table sorted by salience, a red mismatch where the intended entity and the detected one differ, and “637 documents ready” with an estimated cost of $0.00 for the full corpus. The lesson flags the line that reports renamed columns, because a silently renamed column is how errors appear three tools downstream.
Once both files are exported, the lesson takes the profiles straight into the Entity Audit Scorecard to produce Clarity, and shows how to read the metrics behind the score: the dominant entity in titles, H1s and meta descriptions, pages with no dominant entity and pages too diffuse to trust. It also gives a short procedure for checking your chrome filter on your own data, since the right threshold varies from site to site.
The troubleshooting section translates the five errors that most often stop a run — an API not enabled on the key’s project, no billing account, an unrecognised content column, a rate limit partway through and an invalid key — into the action each one needs, and notes that completed documents are kept when a run is resumed, so nothing is paid for twice.
This practical continues Module 2, Diagnostics: Extracting the Truth About Your Site, in The AI-powered Entity Audit, and leads into Lesson 2.4 on using Claude as an entity analyst. It suits learners who want a written checklist beside the tool while they run their own site, and anyone producing the extraction files for a team.
In this lesson you’ll cover: