Refresh job board blog posts with Google Search Console

This workflow requires a Google Search Console connection in Cavuno. First, connect Search Console in your Board settings, then connect Cavuno to your AI client. Cavuno MCP supplies your job board domain, complete published-post inventory, public URLs, current content, and the Board-bound Search Console reads needed to find genuine refresh candidates. You do not need to supply a property, URL prefix, Google credential, or post IDs.

The goal is not to make old articles longer. It is to recover a worthwhile search opportunity by making the right article more useful than the strongest current results without discarding the parts that still work.

Try it

Ask:

Use Cavuno to list all my published blog posts, their publication dates, and their public URLs. Query the connected Google Search Console data for those pages. Compare each post’s recent 28-day window with its best previous 28-day window, exclude posts published or substantially refreshed within the last 90 days, and filter out pages without enough search demand. Rank the largest recoverable traffic losses by lost clicks. Do not change anything yet.

Ask for the evidence behind the strongest candidates:

For each leading candidate, show the recent and peak windows, lost clicks, impressions, click-through rate, average position, lost queries, declining queries, holding queries, and emerging queries. Separate likely content decay from seasonality, falling demand, a sitewide change, a tracking problem, or a title and description problem.

Once you choose a candidate, continue with:

Build a refresh brief for the best candidate using its Search Console evidence and current Cavuno content. Preserve the slug, useful sections, working queries, images, and internal links by default. Then inspect the live search results, identify concrete information gain, and propose an improved outline. Do not rewrite the article merely to make it longer.

Then ask the agent to draft the approved refresh:

Write the refresh section by section, comparing each major section with the strongest current result for that subtopic. Assemble the article, remove repetition, and complete an internal link audit in both directions. Show me the finished article and the exact metadata or links that would change before saving it.

Then, when you are happy with the revision:

Update the post in Cavuno, read it back, and confirm the saved article still has the same slug, public URL, and publication status. Show me the final metadata, links, and key refreshed sections.

What the agent will do

1. Build the post universe. Use Cavuno MCP to discover the primary domain and list every published blog post. It combines each slug with the domain itself to create the public URLs.
2. Measure decay consistently. Query finalized Search Console page data through Cavuno MCP, map it to the Cavuno URLs, and compare each recent 28-day window with the best earlier 28-day window. It excludes new or recently refreshed posts still inside a 90-day ramp and pages without enough demand to justify a refresh.
3. Rank recoverable loss. Prioritise lost clicks rather than the largest percentage decline. A page falling from two clicks to zero is not automatically a better opportunity than a page that has lost substantial qualified traffic.
4. Create a refresh brief. Record the peak and recent periods, lost queries, declining queries, holding queries, emerging queries, recent impressions, click-through rate, position changes, and the likely reason for the decline.
5. Diagnose before rewriting. Separate content decay from seasonality, falling search demand, tracking problems, sitewide changes, intent shifts, and weak titles or descriptions. High impressions with weak click-through rate may call for metadata work rather than a full rewrite.
6. Snapshot the current article. Read the complete Cavuno post and record its title, slug, status, metadata, links, images, and sections. Preserve the public URL and useful existing material unless there is a specific reason to change them.
7. Research the live results. Open the current ranking pages, confirm the search intent and page type, and record what the strongest results include, believe, and omit. Convert those omissions into specific information gain such as current examples, a decision framework, a stronger template, first-party knowledge, or a more precise workflow.
8. Gate the outline. Compare the planned refresh with the strongest results. It should answer the current intent, retain what still works, resolve material gaps, and be clearly more useful—not simply longer.
9. Write section by section. Benchmark each major section against the best current section for that subtopic, revise it, assemble the article, and remove repeated definitions, examples, caveats, and calls to action.
10. Complete the internal link audit. Preserve useful links, remove stale or cannibalising links, add relevant outgoing links, and identify existing Cavuno posts that should link to the refreshed article. Each recommendation includes the source, destination, anchor text, placement, and purpose.
11. Update and verify. Save only the selected post, read it back through Cavuno MCP, and confirm its content, metadata, slug, public URL, and publication state. Search performance should be monitored over comparable future periods rather than judged immediately.

Cavuno MCP can access finalized Search Analytics and Google's indexed URL Inspection result for the connected Board. It cannot browse live search results, run a live URL test, or request indexing; your AI client must supply browser research when the workflow needs it. Search Analytics rows are top rows rather than exhaustive data, so the agent must preserve Google's incompleteness caveats when comparing pages and queries.