Recent work
Diagnosis and recovery, in progress
Case study | Technical SEO, content strategy and AI search visibility
From Blog Traffic to Buyer Traffic: Rebuilding Organic Search for an E-commerce Jeweler
A U.S. custom jewelry brand had strong blog traffic but little of it came from buyers. After Google's March 2026 core update exposed the gap, Corelead diagnosed why and refocused organic search on the pages that sell.
~100,000
possible URLs in a configurable catalog
59%
of organic clicks from non-commercial posts
13 to 5
position move for a category page in a month
2,100
shopper queries mapped to pages
At a glance
- Client
- U.S. e-commerce brand selling custom gemstone engagement rings, wedding bands and fine jewelry
- Services
- Technical SEO, content strategy, search data analysis, migration SEO, AI search visibility, reporting
- Status
- Diagnosis and recovery, in progress
The challenge
The brand's catalog is large and configurable. Roughly a thousand base designs expand into more than 100,000 possible URLs through stone, metal and size options.
SEO scope had been narrowed to blog content, while the money pages relied on paid campaigns and remarketing. In early 2026 the blog drove strong traffic (7,500 to 8,000 organic clicks a month), but product and category pages mostly sat on page 2 or lower, and organic revenue had plateaued.
What we found
- 1
Most traffic wasn't commercial
Astrology and zodiac-style posts supplied about 59% of 12-month organic clicks.
- 2
A step change after the March 2026 core update
Monthly clicks fell from about 7,450 in March to about 2,700 in April, across every blog category at once. Zodiac content accounts for about 71% of the net loss between April and September. Category pages lost about half their clicks.
- 3
Infrastructure was working against the site
The blog sitemap hadn't updated since 2024 because of a conflict with the SEO plugin, and Search Console showed hundreds of pages crawled or discovered but not indexed. Uptime monitoring logged 89 incidents totaling about 151 hours of downtime over 18 months, plus a bot traffic spike of ten times normal volume.
- 4
The catalog had too many URLs
The 353 top-selling models spread across about 4,700 URLs, and together earned only about 4% of organic clicks.
- 5
AI search saw the brand as an astrology site
It appeared in hundreds of AI answers, almost all for non-commercial content, and was absent from AI answers for buyer queries.
- 6
Impressions overstated demand
Cross-checking every priority query against independent search volume caught three queries whose impressions were many times real demand.
AI search visibility: the same problem, one layer up
AI search read the brand the way its traffic looked. It appeared in hundreds of AI answers, almost all for non-commercial content, and was absent from AI answers for buyer queries.
That is why SEO and AI search visibility have to be fixed together. The pages and signals that tell Google what a brand sells are the same ones AI tools draw on. How we approach AI search visibility.
What we did
- 1
Technical audit
Crawled 257 category pages and flagged 129 with title or description problems. Delivered a prioritized rewrite workbook that protected pages already ranking first, implemented by the client's developers.
- 2
Canonical strategy
Designed a quarterly process that indexes only the variant pages with real demand, using two signals: recent sales and search impressions.
- 3
Content
Built a pillar-and-hub program of about 18 long-form buying guides.
- 4
Infrastructure escalation
Replaced the stale sitemap and documented the downtime and bot traffic with evidence for the client's developers.
- 5
Migration protection
A pre-launch checklist for the move to WordPress and WooCommerce covering URL parity, redirects, meta carryover, canonical tags and robots settings. We flagged URL differences on staging and advised against launching in peak holiday sales.
- 6
Data foundation
Combined the Search Console API, SEMrush, DataForSEO, Ubersuggest, Cloro and uptime logs into a keyword-to-page map of about 2,100 shopper queries.
- 7
Reporting
A client dashboard on Corelead's custom SEO platform.
Results to date
- Several non-brand commercial terms reached position 1, and a custom engagement category page moved from position 13 to 5 within a month.
- The infrastructure problems were diagnosed with evidence and escalated.
- Organic clicks remain about 65% below April, mostly from non-commercial content. Recovery now focuses on commercial pages.
Current plan
- 1Map each high-demand search to one owner page (stone wedding band searches of 2,900 to 9,900 monthly searches, currently ranking 15 to 30).
- 2Rewrite titles, descriptions and copy on about 25 existing pages.
- 3Fix internal links, such as stones missing from the homepage.
- 4Consolidate product variants and auto-generate titles and structured data for the top 353 models.
- 5No new pages or articles until the shop pages move.
Lessons
Tools
- Google Search Console API
- Screaming Frog
- SEMrush
- DataForSEO
- Ubersuggest
- Cloro
- UptimeRobot
- GA4
- Looker Studio
- Yoast
- Corelead's custom SEO platform
Is your organic traffic actually buyer traffic?
We check how much of your organic traffic lands on pages that sell, how healthy the site is, and how search and AI tools read your brand.