GSC Opportunity Mining
Search Console is the only dataset that tells you what Google already thinks you're relevant for. The fastest organic wins almost always live in queries you already rank 5–15 for.
Why This Exists
Most audits reach for a third-party keyword tool and ignore the one dataset that reflects reality. Meanwhile page-one-adjacent queries sit there, already earning impressions, needing a title tag rather than a new article.
The bigger miss is decay. A single-period report physically cannot see it — and what you're losing is usually more urgent than what you could gain.
This skill always pulls two periods, separates seasonal from algorithmic from genuine decay, and refuses to recommend new content when the win is fixing a page that already ranks sixth.
How It Works
Pull two periods, always
Via the connected GSC source or a Performance CSV — query and page level, with position, impressions, clicks and CTR, plus the prior comparable period so decay is computable.
Mine striking distance
Queries at average position 5–15 with real impressions, ranked by impressions × (1 − current CTR). Then segment: is the ranking URL the right URL? If not it's a mapping fix, not a content fix.
Find the same-day wins
Compare actual CTR to the expected curve for each position. Ranking well below it means a title/meta problem or someone owning the snippet above you — fixable this afternoon, no content work.
Diagnose cannibalization and decay
Group queries where multiple URLs draw impressions, flag flip-flopping, and classify every declining page as seasonal, algorithmic, or decaying. Mislabeling this wastes months.
Two Minutes To First Output
# 1. Clone the bundle git clone https://github.com/seosdm/seo-skills.git # 2. Connect GSC, or export Queries + Pages from the GSC UI (16 months if offered) # 3. Copy skills/gsc-opportunity-mining/SKILL.md into a Claude Project: Run gsc-opportunity-mining for [property]. Last 3 months vs prior period.
No API keys, no setup. The skill is a SKILL.md file plus a worked example — it runs anywhere Claude does.
Sample Output
Verdict: the headline is the decay, not the upside. Nine pages lost 24% of clicks on one date — a broad algorithmic drop, not stale content, so rewriting them would have wasted a quarter. The 38 CTR gaps are title-tag edits worth more than any new article on the roadmap.
See the full worked example in /examples.
I'll Teach You The Rest
Free skill, real judgment behind it. Want it run on your business, or your team taught to run it themselves?
Coach My Team On This north_eastThe technical floor everything else stands on.
SEO Content BriefBriefs built for extraction, not just ranking.
Backlink & Keyword GapThe referring-domain gap that feeds AI citations.