← All cases

Case · E-commerce · Ukraine · 2025

RAG consultant for e-commerce

Requests/hr at peak
100–500
Over the catalog
RAG
Integration
ERP
The project in facts
FormatConsultant agent in a Telegram widget
KnowledgeRAG over the product catalog
Integrationthe client's ERP
Deploymenton the client's servers
Challenge

Consultants drowned at peak.

A large online store: consultants overloaded with routine product selection and customer education — peaks reach 100–500 requests an hour.

A Telegram-widget recommendation agent with RAG over the catalog, integrated with the client's ERP and deployed on their own servers.

Context

What was in the way

  • Peak loads of 100–500 requests per hour
  • Consultants spent their time on routine product selection
  • Product knowledge lived in the catalog and in heads — not in the sales channel
Capabilities

What the system does

01

Product selection

The agent asks clarifying questions and recommends items from the catalog.

02

Catalog-grounded answers

RAG search over live data — answers grounded in the catalog, no inventions.

03

Unloading the team

The agent closes routine consultations and hands the complex ones to people.

What's inside

How we roll it out.

Discovery

Requests and access analysis

We study historical requests, verify CRM and API access, and score the automation potential.

Pilot

A slice of real traffic

The agent runs on part of the live requests with weekly syncs — real dialogues, not a demo.

Launch

Metrics and scale-up

We evaluate volume, response speed and escalation share — then expand the agent to full traffic.

Results

Before → after

Before
After
Peak hours meant a queue on consultants
The agent closes routine selection
An answer depended on a person being free
An answer right away, at any hour
Product data lived apart from the chat
The catalog and ERP feed the agent’s answers

AI agents for business — 2–3 emails a month

Breakdowns, cases and tools already working inside companies.

No spam. Unsubscribe in one click.