Skip to content
Get Ranked On TopIllustrative case studyContent Engine plan

A fictional business. The figures are a modelled scenario, not a client result.

RATNAVALI GEMS · Coloured gemstone jeweller, 2 showrooms + online · Jaipur

How a Jaipur gem jeweller became the source AI search quotes on emeralds, rubies and sapphires

A Jaipur jeweller had the stones and the expertise, but its website only listed products. We cut that knowledge into guides AI search could quote.

41%

of tracked AI answers to gem-buying questions now cite Ratnavali, up from 4% before the guides.

Content & AI search · 15 months · Asia-Pacific · Content Engine plan

Be₃Al₂Si₆O₁₈ · RI 1.577–1.583 · Mohs 7.5–8

IThe flaw under the loupe

Two showrooms full of answers. A website that only listed prices.

Buyers ask AI search how to tell a real ruby or a lab-grown emerald. With product lists and no guides, the answers cited labs and marketplaces instead.

Ratnavali's gemologists answer the same questions every day, across the counter in Jaipur and on WhatsApp: is this emerald natural, has this ruby been glass-filled, why does one sapphire cost four times another? None of that knowledge was online. The site was a catalogue: 1,140 product pages, 38 words each on average.

So when buyers put those questions to AI search, the answers quoted certification labs, marketplaces and forums. Ratnavali turned up in 4% of them, usually as a bare product link.

Diagnosis

Four readings from the refractometer.

  1. 01

    1,140

    product pages

    Each listing was a photo, a carat weight and a price. No origin, no treatment disclosure, no word on the certificate that came with the stone.

  2. 02

    0

    expert guides

    Not one page explained how to tell natural from lab-grown, what oiling means or how heat changes a ruby's value: the questions buyers ask most before spending serious money.

  3. 03

    312

    buyer questions mined

    From two years of showroom enquiry notes, WhatsApp chats, People Also Ask boxes and AI prompts. They clustered cleanly by stone: emerald, ruby, sapphire.

  4. 04

    4%

    of AI answers cited Ratnavali

    A panel of 240 buyer prompts, run monthly on four AI engines. Labs and marketplaces took 67% of the citations; the jeweller that actually cuts the stones took 4%.

IIThe strategy

Cut the knowledge, facet by facet.

Built on our Content Engine plan: content and AI-search work for businesses whose expertise is the product.

  1. 24 guides planned

    Facet 1

    Buyer-question mapping

    312 real questions grouped into three stone clusters of eight, ranked by how often they came up in the showroom and how badly current AI answers handled them. One question, one guide, one facet.

  2. 8 guides per stone

    Facet 2

    Gemologist-written guides

    Each guide is written with the house gemologist. It opens with a two-sentence answer, then the test, photographs under magnification and what it means for the price. Plain enough to quote, deep enough to trust.

  3. 1,140 products enriched

    Facet 3

    Author & lab schema

    Gemologist profiles with credentials; Article and Product schema linking every stone to its lab report number, origin and treatment, so engines can see who is answering and what backs it up.

  4. Monthly, 4 engines

    Facet 4

    AI citation tracking

    The 240-prompt panel re-run every month on four engines. A guide that was read but not cited got a sharper opening answer, a comparison table or a new photograph, then went back into the panel.

IIIRotate the gem

Every facet is a buyer question. It lights up when AI search cites the answer.

Choose a cluster, turn the stone and move through the 15 months. A dark facet has no guide yet, a cut facet has one, a lit facet is being cited.

10× loupe

Click a facet, or focus the stone and use the arrow keys to turn it.

Be₃Al₂Si₆O₁₈ · RI 1.577–1.583 · Mohs 7.5–8

Month 15 of 15

Brilliance

88%

7 of 8 facets lit

  • No guide yet
  • Guide live, not cited yet
  • Cited by AI search

Under the loupe · Emerald I

Cited by AI search

Buyer question, as typed

natural vs lab grown emerald how to tell

Natural or lab-grown emerald: the tests a gemologist runs

The guide's opening answer

Under a 10× loupe, natural emeralds show irregular three-phase inclusions; lab-grown ones show chevron or veil-like growth patterns. A lab report settles it.

Guide live from month 3

First cited in month 5

By month 15: cited by 4 of 4 engines

IVMonth by month

15 months on the cutting wheel.

  1. Months 1–2

    Audit & question mining

    Crawl of every product page, two years of enquiry notes and chats mined into 312 questions, and the 240-prompt AI panel set up as the baseline.

  2. Months 3–5

    Emerald cluster

    Eight emerald guides written with the house gemologist and photographed under magnification. The first citation arrives in month 4.

  3. Months 6–9

    Ruby cluster

    Eight ruby guides, plus lab report numbers and treatment lines added to every ruby product page.

  4. Months 10–13

    Sapphire cluster

    Eight sapphire guides, including neelam checks for buyers choosing a stone for astrology; author schema and gemologist profiles across the site.

  5. Months 14–15

    Polish

    Guides that were read but not cited get sharper opening answers and comparison tables; the three clusters are cross-linked into one buyer's handbook.

Share of AI buyer answers citing Ratnavali
0%10%20%30%40%50%M0M3M6M9M12M1541%

VResults

A stone that finally catches the light.

4% → 41%

AI buyer answers citing Ratnavali

240 buyer prompts × 4 engines, every month

0102030405041%READING

21 of 24

Expert guides cited by AI search

The three still dark are queued for a rewrite

3,100 → 11,900

Organic sessions / month

Monthly, from month 0 to month 15

38 → 96

Qualified enquiries / month

Showrooms and online, per month

3896
Who AI search cited, before and at month 15

Modelled scenario for a fictional business — not a client result. Rankings, traffic and AI citations are never guaranteed; see our Terms.

Get a plan like this

Is your best expertise stuck behind the counter? Start with the Content Engine plan, or tell us which questions your customers keep asking.

← All case studies