STRATAI™
HOME›CASE STUDIES›Mid-sized Jewellery Manufacturer (India)
JEWELLERY MANUFACTURINGMID-SIZED JEWELLERY MANUFACTURER (INDIA)

AI CATALOGUE & ORDER INTELLIGENCE SYSTEM: REDESIGNING INFORMATION FLOW AT A JEWELLERY MANUFACTURER

AI catalogueOrder intelligenceJewellery manufacturingInformation architectureRetainer modelAI searchManufacturing AIIndia
64,000
ACTIVE DESIGNS
Enriched with AI-extracted metadata and enhanced imagery
7 Days
SALES-TO-PRODUCTION LAG
Current lag on a confirmed order; client projects under 1 day
2 Sec
SEARCH TIME
Mobile AI search vs 2-3 days of manual ERP filtering and PDF
₹90,000
MAX GOLD LOSS PER RETURN
Per returned 100-piece order at ₹15,000/g and 3% remelting loss
OVERVIEW

A 64,000-design jewellery catalogue ran on PDF printouts and WhatsApp, causing a 7-day sales lag and costly gold remelting losses. StratAI is redesigning the information flow with AI-enriched data, 2-second mobile search and structured order capture feeding production via API.

KEY TAKEAWAYS
01The problem was information flow, not technology or people: incomplete specifications at every step caused the 7-day lag and remelting losses.
02AI read 64,000 design images to extract structured metadata and enhance imagery, turning a thin database into a searchable asset.
03Mobile search returns specification-complete results in under 2 seconds, replacing 2-3 days of manual ERP filtering and PDF handling.
04Category-specific order capture sends fully populated job orders to production via API, removing floor-level assumptions.
05Results are projected by the client (about ₹3 Crore additional gross margin per year) and will be updated from Month 4-6.
THE CHALLENGE

A 64,000-design catalogue managed through PDF printouts and WhatsApp messages. Every design carries 20+ variables, yet specifications were lost at every step between buyer inquiry and production, causing a 7-day sales-to-production lag on confirmed orders and Rs 45,000 to Rs 90,000 in gold remelting losses per returned order.

WHAT WE BUILT

StratAI redesigned the information flow in three layers: (1) AI enrichment of all 64,000 designs, extracting structured data and enhancing imagery; (2) an AI-powered mobile search that returns matching designs in under 2 seconds; (3) structured, category-specific order specification capture that passes complete job orders to production software via API.

SUMMARY

A 64,000-design jewellery catalogue ran on PDF printouts and WhatsApp, causing a 7-day sales lag and costly gold remelting losses. StratAI is redesigning the information flow with AI-enriched data, 2-second mobile search and structured order capture feeding production via API.

KEY TAKEAWAYS
  • →The problem was information flow, not technology or people: incomplete specifications at every step caused the 7-day lag and remelting losses.
  • →AI read 64,000 design images to extract structured metadata and enhance imagery, turning a thin database into a searchable asset.
  • →Mobile search returns specification-complete results in under 2 seconds, replacing 2-3 days of manual ERP filtering and PDF handling.
  • →Category-specific order capture sends fully populated job orders to production via API, removing floor-level assumptions.
  • →Results are projected by the client (about ₹3 Crore additional gross margin per year) and will be updated from Month 4-6.
Jewellery Manufacturing India Retainer Model Engagement In Progress

How StratAI is redesigning the information architecture of a mid-sized jewellery manufacturer, and what it means for every manufacturer.

The Engagement at a Glance

ClientMid-sized jewellery manufacturer, India
Suppliers toGRT, Joyalukkas, Thangamayil, ABG and major wholesalers across India
Catalogue size64,000 active designs
Customers200+ wholesale buyers, 40+ repeat accounts
EngagementAA-SaaS Retainer, StratAI
StatusSolution design complete. Build underway. Results tracking from Month 4.

The problem: a 64,000-design catalogue managed through PDF printouts and WhatsApp messages, causing a 7-day sales drag on confirmed orders and ₹45,000–₹90,000 in gold losses per returned order.

01 — The Complexity Nobody Sees From the Outside

From the outside, a mid-sized jewellery manufacturer supplying India's largest retail chains looks like a craft business. From the inside, it is one of the most variable, data-intensive manufacturing operations in any sector. Every single design in the catalogue is defined by 20+ variables:

  • Make type: Casting or handmade (Asari)
  • Metal purity: 18k, 22k
  • Category: Ring, Earring, Necklace, Bangle, Bracelet, Chain
  • Sub-category: Ladies Finger Ring, Hanging Earring, Choker Necklace
  • Design collection: Birds, Temple, Florals, Geometric
  • Sub-collection: Peacock, Lotus, Kaveri
  • Enamel / stone: Yes or No per design
  • Finish: Matt, Polish, Antique
  • Category accessories: Earring: screw or push back. Necklace: end cup, S-hook, back chain. Ring and bangle: size or diameter.
  • Weight: Per piece, per component, custom ranges

02 — How a Simple Order Becomes a 7-Day Ordeal

A wholesale buyer, with an established relationship and a clear requirement, places an inquiry. What follows is a process designed to lose time and information at every step.

Buyer inquiry
WhatsApp, voice note or in person. No formal specification captured. Information lost: the requirement is verbal and nothing is documented.
2–3 days
Sales filters the ERP
Sales manually filters the ERP, selects ~130 designs, generates a PDF and sends it to the buyer. Information lost: no sub-collection, enamel, stone, accessory or weight detail in the PDF.
2–3 days
Buyer reviews the PDF
The buyer prints the PDF, physically strikes out unwanted designs, then photographs and sends it back. Information lost: approval is ambiguous and specification gaps remain.
1–2 days
Order processing
The job order is raised from a WhatsApp message and the struck-off PDF. Information lost: accessory specs, sizes and finish details are all incomplete.
Ongoing
Production starts
The craftsman makes judgment calls on missing specifications. Information lost: floor-level assumptions replace buyer specification.

Total elapsed time: 7 days, for a buyer with a confirmed account, a clear product requirement and an existing relationship. Every day of that delay is a day the manufacturer's position with their wholesale customer weakens.

Industry-standard gold remelting loss is 1.5%–3%. In one recent return, 40% of an order came back due to specification mismatch: ₹30,000 lost in two days of production work. At ₹15,000/g and 3% remelting loss, a returned 100-piece order can cost up to ₹90,000.

03 — Three Layers. One Information Architecture.

StratAI's diagnosis: this is not a technology problem or a people problem. It is an information flow problem. Every breakdown, whether the 7-day lag, the remelting loss or the production floor stress, traces back to information that was incomplete, siloed, or captured in the wrong format at the wrong time. The solution addresses information flow at three layers.

Layer 1 · Database Intelligence

The existing database held thin data and inconsistent imagery, with many images shot of wax prototypes rather than finished gold. StratAI deployed AI to read every image and extract structured data: design collection, sub-collection, enamel presence, stone setting, finish type, and component weight read directly from the weight stamp in each image. Images are enhanced for angle, lighting and consistency so buyers see finished gold, not wax. The result: 64,000 designs with rich, accurate, complete metadata and buyer-ready imagery.

Layer 2 · AI-Powered Mobile Search

A mobile application built on the enriched database. A buyer types, speaks or filters: "2-gram peacock ladies finger ring, 22 karat." All matching designs appear in under 2 seconds, with complete specifications, enhanced imagery and category-specific accessory options. The search that took the sales team 2–3 days of manual ERP filtering and PDF generation now takes the buyer 2 seconds, independently, on their phone, built for how wholesale buyers actually work: on the move, between trade shows and appointments.

Layer 3 · Order Specification Capture

Every specification is captured at the point of selection through structured, category-specific input fields. An earring order prompts for screw back or push back and size. A necklace prompts for end cup, S-hook, back chain and chain length. Once confirmed, the complete order specification passes directly into the production software via API, so the production floor receives a job order where every field is populated, eliminating assumption-based decisions and the remelting losses they cause.

How StratAI Works

Every engagement begins with a 3-month deep on-ground study of the client's information architecture: systems, data, people and breakdown points. System design follows (1 month). Development scope varies by complexity. StratAI operates exclusively on a retainer model, because compounding knowledge of the client's business, relationships with their team and continuous system evolution cannot be delivered on a project basis. StratAI currently works with 10+ manufacturers on multi-year retainer engagements.

04 — Projected Impact

Results are pending and tracked from Month 4 per StratAI's 6-Month P&L Horizon framework. The projections below are the client's own estimates, based on their deep knowledge of their wholesale network. The use case identification, solution design and implementation are StratAI's.

Sales cycle, confirmed account
Today: 7 days from inquiry to production start. Projected: less than 1 day.
Buyer catalogue experience
Today: PDF printout and a 2–3 day wait; 98% of buyers do not use the existing web app. Projected: mobile AI search with 2-second results and full specification visible.
Order specification completeness
Today: incomplete, with WhatsApp and a struck-off PDF as the production reference. Projected: 100% structured, every field populated before production begins.
Remelting loss from returns
Today: ₹45,000–₹90,000 per returned order (100 pcs, 2g, at ₹15,000/g). Target: near zero, with specification mismatch eliminated at source.
Additional wholesale orders per year
Today: orders constrained by a slow, friction-heavy buying process. Projected by the client: 100 kg of additional gold orders after full wholesaler onboarding.
Gross margin impact per year
~₹3 Crore additional gross margin (100 kg × ₹15,000/g × 2–3% margin), per the client's projection.

Results tracking begins Month 4. Full P&L impact is expected from Month 6 onwards. This case study will be updated with live results at that stage.

05 — The Manufacturing Systems Principle

Strip any manufacturing operations problem down to its core and you will find the same three failures: (1) Information is incomplete, inconsistent, or trapped in silos. (2) Goods flow through a process built around missing information, building in rework, return and waste from the start. (3) People coordinate manually across the gaps, absorbing time, errors and stress that should never have been theirs to carry. Fix the information flow, and the goods flow and people coordination problems largely solve themselves.

This case study is published while the engagement is in progress. StratAI publishes in-progress work deliberately, because the depth of problem diagnosis and solution design is visible before results are. Results will be added at Month 6.

Talk to StratAI

If your manufacturing business has a complex information flow problem, in any sector and at any scale, talk to StratAI.

→ stratai.io/contact
FREQUENTLY ASKED QUESTIONS

What problem did StratAI solve for the jewellery manufacturer?

A 64,000-design catalogue managed through PDF printouts and WhatsApp caused a 7-day lag between a confirmed order and production start, and incomplete specifications led to returns costing Rs 45,000 to Rs 90,000 in gold remelting loss each.

How does the AI mobile search work?

A buyer types, speaks or filters a requirement such as a 2-gram peacock ladies finger ring in 22 karat, and all matching designs appear in under 2 seconds with full specifications, enhanced imagery and accessory options.

Are the results final?

No. The engagement is in progress. Solution design is complete and the build is underway. The projected impact reflects the client's own estimates, results tracking begins in Month 4, and the case study will be updated at Month 6.

Does this approach apply to other manufacturers?

Yes. StratAI's framework treats most manufacturing problems as information flow problems, so the same approach applies to any sector with a complex catalogue or order specification process.

BY THE NUMBERS
64,000
ACTIVE DESIGNS
Enriched with AI-extracted metadata and enhanced imagery
7 Days
SALES-TO-PRODUCTION LAG
Current lag on a confirmed order; client projects under 1 day
2 Sec
SEARCH TIME
Mobile AI search vs 2-3 days of manual ERP filtering and PDF
₹90,000
MAX GOLD LOSS PER RETURN
Per returned 100-piece order at ₹15,000/g and 3% remelting loss
TAGS
AI catalogueOrder intelligenceJewellery manufacturingInformation architectureRetainer modelAI searchManufacturing AIIndia
WANT SIMILAR RESULTS? →
FREQUENTLY ASKED QUESTIONS
What problem did StratAI solve for the jewellery manufacturer?+
A 64,000-design catalogue managed through PDF printouts and WhatsApp caused a 7-day lag between a confirmed order and production start, and incomplete specifications led to returns costing Rs 45,000 to Rs 90,000 in gold remelting loss each.
How does the AI mobile search work?+
A buyer types, speaks or filters a requirement such as a 2-gram peacock ladies finger ring in 22 karat, and all matching designs appear in under 2 seconds with full specifications, enhanced imagery and accessory options.
Are the results final?+
No. The engagement is in progress. Solution design is complete and the build is underway. The projected impact reflects the client's own estimates, results tracking begins in Month 4, and the case study will be updated at Month 6.
Does this approach apply to other manufacturers?+
Yes. StratAI's framework treats most manufacturing problems as information flow problems, so the same approach applies to any sector with a complex catalogue or order specification process.
WANT SIMILAR RESULTS?

LET'S TALK.

Tell us what you are working on and we'll design the right approach.

WE RESPECT YOUR PRIVACY AND WILL NEVER SHARE YOUR INFORMATION.

Written by
Palaniappan SN
Palaniappan SN
www.linkedin.com/in/palaniappan-sn-b10820108
Co-Founder, StratAI · MBA, IIM Bangalore · BE (Mechanical), PSG Tech

Palaniappan SN is a Business Strategy Consultant who has spent his career at the intersection of business strategy and operational reality — working across management levels from the boardroom to the shop floor to understand where organisations actually win and lose. His conviction is simple: AI should never be an experiment. It should be an advantage. That belief is the foundation of StratAI's AI Advantage Systems methodology — built not from technology-first thinking, but from the ground up, with the discipline to walk away from projects where the conditions for success don't exist.

MORE CASE STUDIES
LAKSHMI KRISHNA NATURALS
AI-Powered Customer Intelligence & Sales Automation for E-Commerce Growth
19
GLOBAL MARKETS
32+
KEY CUSTOMERS
GERMAN BUYING HOUSE (TIRUPUR, INDIA)
From Shop Floor to P&L Impact: AI Systems for a German Buying House in Tirupur
3:45 → 1:45
MEASUREMENT TIME
1.5 hrs → 0
POST-SHIFT ENTRY
CONSUMER MEDICAL DEVICES COMPANY (INDIA)
The Power of the Domain-First Principle in AI Transformation
7-8
MARKETPLACE SOURCES
Month 4
ENGAGEMENT STATUS