Traffic Drop Diagnosis from Google Search Console - SEOmatic

Diagnose why organic traffic dropped on the connected site from real Search Console data: separate seasonality from real loss, find which pages and queries lost the clicks, check indexing and competitors, and rank what to fix first. Use when the user says traffic, clicks or rankings fell, or asks what happened to their SEO. Runs on SEOmatic, the SEO agent that works from the site's real Google Search Console data; nothing goes live until the user approves it.

Traffic drop diagnosis

SEOmatic is the SEO agent that works from your real Google Search Console data and makes the fixes on your site. Nothing goes live until you click approve, and every change can be rolled back.

"Traffic dropped" has five common causes and they need opposite fixes. This skill finds which one it is, with numbers, before anyone touches a page. Tools are scoped to the connected site, so never ask for a domain.

Prerequisite

Search Console must be connected. If a tool reports it is not, give the user this link and stop: https://app.seomatic.ai/connect/gsc - it is one browser step, then re-ask. Do not substitute guesses for GSC data.

1. Confirm the drop is real

gsc_insights action seasonality first. It compares 16 months year over year and says whether this is a real change or the same dip as last year. A seasonal dip is not a problem to fix, and saying so is the most useful answer you can give.

Then gsc_performance action compare_periods (28 days against the 28 before) and action trend to find the day it started. A cliff on one day points at a change (a deploy, a migration, an update). A slow slope points at competition or content decay.

GSC data lags 2-3 days. If the drop is in the last 48 hours, check gsc_performance action fresh and say the numbers are provisional.

2. Find where the clicks went

• gsc_performance action folder_rollup: is it one section of the site or everything? Sitewide points at technical or algorithmic causes; one folder points at that folder.
• compare_periods on pages, then on queries: the 10 biggest losers by clicks lost, with position before and after.
• Position steady but clicks down: demand fell, or a SERP feature took the click. Check serp_competitors action features on the top lost queries.
• Position down: someone outranked you, or the page changed.

3. Rule out the technical causes

gsc_indexing action batch_inspect on the top losing URLs. Look for "crawled - not indexed", a canonical pointing elsewhere, noindex, or a fetch error. gsc_insights action sitemaps for sitemap errors. One noindex on a template can take out a whole section overnight.

site_pages action crawl on the two biggest losers: did the title, H1 or body change? A redesign that dropped the copy is a common, fixable cause.

4. Check it is not just measurement

If GA4 is connected, traffic_analytics action sources for the same window. If organic sessions in GA4 held while GSC clicks fell (or the other way round), the problem may be tracking, not SEO.

5. Give a verdict, then the fixes

Lead with one sentence: which cause, how sure you are, and the evidence. "Clicks fell 34% from the 3rd. 80% of the loss is the /blog/ folder, and 6 of its top 10 pages now show 'Crawled - not indexed'." Then the fixes, ranked by clicks recovered. Say what you could not check.

Offer to stage the work

Staging needs a paid SEOmatic plan (the key carries the agents:act scope). It costs no credits, runs nothing, and always waits for the user's approval. On a free key the staging actions are locked and say which plan includes them: describe the specific fix instead so the user can pass it on.

Check task_manage action list first so you do not stage a duplicate. Then task_manage action create with up to 5 tasks: content_refresh for decayed pages, ctr_fix (needs queries) where position held but CTR fell, index_request or technical_fix for indexing problems. Each task needs targetKey (the full page URL, or a topic for content work), a title that states the goal in the user's words ("Get 'pricing calculator' onto page 1", never "Optimize meta tags"), and a rationale with the numbers behind it. Quote cost from each task's estimatedCost.label.

• Tell the user the cards are staged and where to approve them.
• Never approve on the user's behalf. Approval happens in SEOmatic, or through task_manage action decide only after the user says yes to that specific task.
• Approved auto-executable tasks start within seconds: read the result back with task_manage action get before you claim anything changed.

What not to do

• Do not blame "a Google update" without evidence. Name it only if the drop date and the pattern match, and say it is a hypothesis.
• Do not call a seasonal dip a drop.
• Do not invent numbers. Every figure comes from a tool result, with its date range quoted exactly.