Sep 20,2026 
By Mansoor Bhanpurawala 
TL;DR
Over three months, this UK jewellery retailer’s content went from roughly 400 AI Overview impressions a day to daily spikes above 1,400 a jump confirmed independently by both Google Search Console and Ahrefs. Organic clicks rose from 3,550 to 5,060 over the same comparison window and average position improved from 23.7 to 17.2.
Not every number moved the right way: click-through rate fell, and online sales conversion remains under 1%. Both are addressed honestly below, not left out.
The client & problem
An independent, multi-generation family jewellery retailer in the UK, trading on Shopify and specialising in bespoke engagement and wedding rings, remodelling, and repairs. High-consideration, high-price-point purchases the kind of product people want to see in person before they commit.
At the start of the engagement (April 2025), the business had:
- Very little organic traffic and effectively no online sales
- No meaningful visibility in search or AI-generated answers for the queries that actually matter informational or transactional
- No way to connect website activity to in-store visits, calls, or WhatsApp enquiries
- Site-level problems undermining everything downstream: a broken ring-builder tool, contact form validation errors, a poor mobile experience, and unreliable conversion tracking
Worth knowing
No single clean “before” snapshot exists for this engagement analytics tracking wasn’t fully in place at the start, and a Google Merchant Center bug corrupted parts of the Search Console data mid-campaign. Where a clean before/after isn’t available, the numbers below reflect a directional trend, not a manufactured multiplier.
What We Did
- Resolved core page speed issues and optimised category and product pages, prioritising the engagement and wedding ring collections
- Optimised the Google Business Profile listing
- Set up and cleaned the Shopify-to-Google-Merchant-Center product feed
- Added Product schema markup for Google Shopping eligibility and for readability by LLMs (ChatGPT, Claude, Gemini)
- Rebuilt header and footer navigation and mapped orphaned pages back into the site structure
- Ran a competitor gap analysis and built a content plan around what it surfaced
- Wrote answer-first blog content aimed at AI Overview and LLM citation, not just classic ranking
- Built UK-relevant backlinks through business listings and guest posts within the jewellery niche
Ten pieces of content from the gap analysis are live, covering monthly birthstone guides, engagement and wedding ring buying guides, gold-purity comparisons, and gemstone-specific ring guides.
Timeline
April 2025 — Engagement begins with a foundational technical and UX audit
2025–2026 — Content build-out, schema implementation, backlink acquisition, GMB optimisation, ongoing
Mid-2026 — A Merchant Center tracking bug affecting Search Console reliability is identified and resolved
Now — Active, ongoing engagement
Results
Search & AI visibility
| Metric | Value | Source |
|---|---|---|
| AI Overview impressions | 65,500 in 3 months | Google Search Console |
| AI-search referral visits | 469 all-time | Ahrefs |
| Organic search visits | 15,390 (38.2% of total traffic) | Ahrefs |
| Average position | 23.7 → 17.2 | Google Search Console |
| Organic clicks (6-month comparison) | 3,550 → 5,060 | Google Search Console |
What I’d do differently
The trade-off this build didn’t manage
Navigational traffic fell 50% and transactional-intent traffic fell 60% between July 2024 and February 2026, even while informational and commercial traffic grew substantially. The programmatic stock-ticker build did its job, but it likely pulled internal link equity and search-console classification away from the site’s core transactional and navigational demand and that split was never given its own tracked target. A site this dependent on programmatic long-tail content needs an internal-linking strategy that explicitly protects its conversion pages, not just one that feeds the hub-and-spoke blog.
The “stagnant Authority Score” story also rests on only two verified points fifteen months apart 49 in July 2024, 51 in October 2025 that happen to land in the same band. That’s suggestive of a flat period, not proof of a continuously flat one; a tracked monthly baseline from day one would have made the claim airtight instead of inferred from two bookends.
Read this honestly
92.8% of AI-search referral traffic in this account comes from ChatGPT alone. That’s real, measurable growth in a channel that barely existed for this site a year ago — and it’s also concentration in a single engine, not evenly spread GEO performance yet.
Local visibility & direct contact (Google Business Profile)
Metric Value Year-on-year Business Profile interactions 901 +16.7% Calls 145 +51.0% Chat clicks 27 +237.5% Website clicks 359 +24.2% Profile views 4,657 -2.3% Search appearances 291 -61.5% Tension worth naming
Contact actions (calls, chats, website clicks) are all up sharply. Profile views and search appearances are both down over the same window. More people who find the listing are acting on it — but fewer people are finding it in the first place. That’s a Google Business Profile visibility question still being worked on, not a settled win.
Direct contact signals from the website
WhatsApp and phone-call clicks make up a genuinely large share of tracked outbound activity from the site ahead of email or social. That supports a real pattern for this kind of premium purchase: visitors browse the site, then choose to call or message rather than check out online. It’s good evidence of intent. It doesn’t prove those conversations turned into sales; that link isn’t currently tracked, and the case study doesn’t claim otherwise.
Revenue
Online sales through Shopify: £6,047 across the engagement window, 13 orders.
Caveat: read before publishing
This figure includes a short period of test/dummy orders created before the engagement began. We haven’t isolated the exact contribution of that test data with a clean re-export. Report this number with that caveat attached, or replace it with a corrected figure once one is available.
Online conversion rate sits at roughly 0.02% (13 orders against 65,947 sessions), which is low, and confirmed by the client as an accurate reflection of current performance, not a tracking error.
Removed
An in-store/offline sales figure was originally part of this engagement’s reporting. It’s been removed from this case study: it was based on customers self-reporting they saw products online before buying in person, with no transaction-level proof or tracked attribution. If it’s worth including at all, it belongs in the next section as an unmeasured channel, not as a dollar figure next to the verified numbers above.
What I’d do differently
- The funnel here is inverted: strong, growing top-of-funnel visibility through AI Overviews and informational content, but a conversion rate that hasn’t moved. That gap should have been instrumented from day one; a tracked promo code or a simple “mention this when you call” line would have closed the online-to-in-store attribution gap instead of leaving it as guesswork.
- AI-search visibility is currently a one-engine story: Real, but it needs to diversify beyond ChatGPT to count as durable GEO performance.
- The single largest driver of clicks in the broader Search Console data is still the branded query. Non-branded ranking growth is real, but smaller than the topline click numbers suggest.
- Google Business Profile search appearances dropped over 60% even as engagement rose. That needs its own investigation, not a footnote.
- Tracking hygiene cost real-time on this engagement: a Merchant Centre bug and a period of test orders both had to be identified and worked around mid-campaign. Every number in this case study would have been faster to produce, and easier to trust, with clean baseline tracking in place from the start.
In the client’s words
Mansoor Bhanpurawala is an AI SEO Consultant with over 14 years of hands-on experience driving organic search growth for enterprise SaaS brands and high-traffic publications. He holds a Bachelor’s Degree in IT and is certified in Technical SEO and Google Analytics. Throughout his career, Mansoor has audited more than 500 websites, managed content teams, and scaled multiple sites to over 500,000 monthly visitors. His technical insights have been featured in Mid Day, Yahoo, Dailyhunt, and Upcity. You can review his full case studies at digitalmansoor.com or connect with him directly on LinkedIn.