Skip to content
Diamond Revenue OS for B2B diamond sales
IQ DiamondsLangGraph · CRM · RAG · Python

DiamondRevenueOSforB2Bdiamondsales

Sales run on managers’ memory and scattered spreadsheets — part of the base sits untouched for months, price and discount are set by eye, and deals leak through forgotten follow-ups. The AI keeps 100% of the base under regular touches and prepares a quote in minutes (pricing decisions stay with a human), cutting lost deals and lifting average margin.

Brief

An AI Revenue OS for B2B diamond sales (IQ Diamonds) — 5 modules from Product to Revenue Intelligence. The AI runs leads, touches, qualification and quotes; pricing decisions stay with a human.

Context

The cost of manual work

Sales run on managers’ memory and scattered spreadsheets — part of the base sits untouched for months, price and discount are set by eye, and deals leak through forgotten follow-ups.

What AI changes

The AI keeps 100% of the base under regular touches and prepares a quote in minutes (pricing decisions stay with a human), cutting lost deals and lifting average margin.

Before → After

Operating model

Before
  • Sales run on managers and spreadsheets
  • Price and discount are set by hand
  • Part of the base sits untouched for months
6 weeks
After
  • +The AI runs leads, touches and qualification
  • +A quote is prepared in minutes
  • +Pricing decisions stay with a human
What we did

How we built it

01

Diagnostic

We mapped the sales process: the base, touches, pricing. We locked the outcome metrics: share of the base under touches and quote speed.

02

Prototype on real data

The first working loop on the real deal base — from leads to quotes — within 1–6 weeks.

03

Production & support

We took 5 modules from Product to Revenue Intelligence into production: 100% of the base under touches, quotes in minutes, pricing decisions with a human.

Results

What came out

minutes

per quote

100%

of the base under touches

follow-up conversion

Project team
IG

Igor Golikov

Project lead

VK

Vitaly Kust

Tech lead

Next case

AI for the Fortuna Group mall network

The mall portfolio is managed by hand and reporting arrives weeks late — a tenant’s dip or a traffic drop is spotted after the money is already lost, and without e-document workflow paperwork bogs down. The AI brings the whole portfolio onto one screen and catches deviations the same day, turning after-the-fact reports into real-time management and protecting the network’s revenue.

AI for the Fortuna Group mall network