Earnings: Search Console clicks by article and by source

Earnings answers one question: which source of articles is paying off. It shows the clicks and impressions your published articles earn in Google Search, grouped by where each article came from. It counts outcomes only. It shows no average position, and it is not rank tracking.

What Earnings shows#

• Clicks by source: one bar per source, radar, research, idea and manual, totalling the clicks of the articles from that source. The colors match the source chips on the Desk and in the Library.
• Per-article: one row per article, with Article, Source, Project, Impressions and Clicks. The title links to the article's detail page.

A row marked estimated match was matched to its live page by a similar path, not by its exact URL, so its numbers may belong to a neighboring page.

The screen covers the projects of every organization you belong to.

The sample preview#

Until the first real numbers arrive, Earnings shows sample data so you can see the layout. It says so in three places: a banner reading Sample preview — not your data., a watermark on the chart, and a Sample tag on every row. The sample rows are fixed illustrations, not estimates for your site. The screen switches to your own numbers as soon as one real row exists.

Where the data comes from#

The numbers come from Google Search Console, through a read-only connection you make per project under project settings, on the Search Console page, with Connect Google Search Console. It is a separate Google grant, not your VarynForge login, and it must be the Google account that owns the site's Search Console property. Project settings covers the page.

Once connected:

• A project syncs when its newest snapshot is more than seven days old, so numbers arrive about weekly. The check runs daily.
• Each sync reads up to three days ago, because Search Console data lags by about two days. An article's first sync goes back 28 days.
• An article is matched to Search Console pages by the live URL recorded when it was published. An article marked published without a URL has nothing to match and stays off this screen.
• Each article's clicks and impressions are the sum of its snapshots, which never overlap.

The sync reads Search Console's own reports. It does not scrape search results.

What you can do here#

Earnings is a read-only screen. Its controls are the article links and the chart's hover detail. To get an article onto it, give the article a live URL: Confirm publish on the Desk, or the matching tool below. Publishing and attestation covers what a publish records.

The same work over MCP#

No MCP tool returns these numbers; Earnings and the article's detail page are the only places they appear. What your agent controls is the input. mark_article_published records the live URL that Search Console matching keys off, with publishedAt when the article went live earlier than the call. update_article_status set to published records no URL, so that article earns nothing here. For an article a CMS will publish later, the agent sets scheduledFor with update_article_status, and publish detection records the URL when the page appears.