--- name: google-search-console-content-decay description: "On the 1st of each month: pages that lost a quarter or more of their clicks over the last 28 days, checked against last year so seasonal dips stay out. Use when the user asks for this, or when it runs on a schedule." --- # Search Console content decay On the 1st of each month: pages that lost a quarter or more of their clicks over the last 28 days, checked against last year so seasonal dips stay out. Content decay is slow enough that week-over-week reports never catch it. A guide loses 4% a week for two months, and by the time anyone looks it has lost a third of its traffic. The causes vary (a competitor's fresher page, an AI Overview, an outdated year in the title), and so do the fixes. The agent compares the last 28 days with the 28 before, checks the same windows last year to rule out seasonality, and names the queries each decaying page lost. What you get is a refresh list with the reasons attached. ## Two ways to get the data Use the API or MCP route when you have an MCP server, a CLI or an API token for the app. Otherwise, if you can drive a browser the user is signed in to (Claude in Chrome, a Playwright or Chrome DevTools MCP server), use the browser route. Say which one you used. ## Route A: API or MCP server ### What you need - Read access to a Search Console property with at least 13 months of data, for the seasonality check. - Search Console API access for the agent: an MCP server, a connector, or OAuth credentials with the `webmasters.readonly` scope. If something is missing, say what and stop; do not invent numbers. ### Steps 1. **Pull two 28-day windows** (Google Search Console): Call `searchanalytics.query` with `dimensions: ["page"]` for the 28 days ending 2 days ago and for the 28 days before that, `type: "web"`, keeping clicks, impressions and position. 2. **Check the same windows last year** (Google Search Console): Pull both windows again shifted back 364 days, which Search Console's 16 months of history allows. A page whose clicks fell by a similar share last year is seasonal; leave it out. 3. **Pick the decaying pages**: Keep pages that lost at least 25% of their clicks, and at least 50 clicks in absolute terms, that did not pass as seasonal. 4. **Find the queries that went** (Google Search Console): For each page, query `dimensions: ["query"]` filtered to that page for both windows, and list the queries with the biggest click losses with their old and new position. 5. **Name the likely cause**: Position down means you were outranked: refresh the content. Position steady but CTR down means something new sits above you, often an AI Overview or a snippet: rework the title and the answer at the top. Impressions down at a steady position means demand fell. ## Route B: the browser ### What you need - Signed in to Google with at least Restricted access to a Search Console property that has 13 months of data for the year-ago check. If something is missing, say what and stop; do not invent numbers. ### Steps Start at https://search.google.com/search-console/performance/search-analytics. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears. 1. **Compare the two 28-day windows**: Pick the property, open Performance > Search results, click the Date filter, choose Compare, `Compare last 28 days to previous period` and Apply. Turn on all four metrics and open the Pages tab. 2. **Pick the decaying pages**: Sort by the Clicks Difference column, lowest first. Keep pages that lost at least 25% of their clicks and at least 50 clicks, and note their position in both periods. Export > Google Sheets when the table is long. 3. **Rule out seasonality**: Change the Date filter to a Custom comparison: the same two 28-day windows shifted back 364 days. A page that fell by a similar share then is seasonal; leave it out. 4. **See which queries went**: Return to the current comparison, click a decaying page (adds a Page filter) and open the Queries tab. The Difference column shows the queries that lost the most clicks, with position for both periods. 5. **Name the cause and the first refresh**: Position down means outranked: refresh the content. Position steady with CTR down means something new above you, often an AI Overview: rework the title and the answer at the top. Impressions down at a steady position means less demand. Note: The Pages table caps at 1,000 rows, which covers the pages that can lose 50 clicks on most sites. The year-ago comparison is one more date change in the same report, so this skill reads well in the browser. ## The report - **Clicks lost** across decaying pages over the last 28 days. - **Decaying pages**: clicks now and before, the change, and position now and before. - **Queries behind each loss**, with old and new position. - **Likely cause** per page: outranked, pushed down by a results-page feature, or lower demand. - The page to refresh first. ## Post it to Briefkin When `BRIEFKIN_TOKEN` is set, or `~/.config/briefkin/env` exists (the installer from https://briefkin.dev/docs writes it), also send the result to the user's Briefkin board as one card. Keep the `item_key` the same on every run so the card updates in place, use a new `event_id` per run, and put today's real numbers in place of the example ones. If the request fails, say so once and carry on. ```bash [ -n "${BRIEFKIN_TOKEN:-}" ] || . ~/.config/briefkin/env curl -sS -X POST "${BRIEFKIN_URL:-https://briefkin.dev}/v1/events" \ -H "authorization: Bearer $BRIEFKIN_TOKEN" \ -H "content-type: application/json" \ -d @- <<'JSON' { "v": 1, "event_id": "google-search-console-content-decay/", "item_key": "google-search-console-content-decay", "project": "example.com", "source": "Search Console content decay skill", "kind": "update", "title": "4 pages lost a quarter of their clicks", "summary": "Four pages lost **2,110 clicks** over the last 28 days against the 28 before. None of them dipped like this a year ago, so it is not the season.", "widget": [ { "type": "Table", "props": { "columns": [ "Page", "Clicks", "Before", "Change", "Position" ], "align": [ "left", "right", "right", "right", "right" ], "rows": [ [ "/templates/invoice-excel", 2240, 3050, "-27%", "2.8 → 3.6" ], [ "/blog/best-invoicing-software", 1120, 1870, "-40%", "3.2 → 5.9" ], [ "/guides/vat-invoice-requirements", 410, 690, "-41%", "6.4 → 9.1" ], [ "/blog/invoice-vs-receipt", 640, 910, "-30%", "2.1 → 2.4" ] ] } }, { "type": "Callout", "props": { "text": "/blog/invoice-vs-receipt held position 2 yet lost 30% of its clicks. Look for a new AI Overview on its main query.", "tone": "warn" } } ] } JSON ``` Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report. ## Run it on a schedule Suggested: 1st of the month at 10:00 (`0 10 1 * *`). From cron, with the agent signed in on that machine: ```bash 0 10 1 * * claude -p "Run the google-search-console-content-decay skill" # Codex: codex exec, Gemini CLI: gemini -p, Copilot CLI: copilot -p, OpenCode: opencode run ``` Or use the agent's own scheduler (Claude Code routines, Codex automations, Claude or ChatGPT scheduled tasks): https://briefkin.dev/skills/google-search-console-content-decay#schedule Source: https://briefkin.dev/skills/google-search-console-content-decay