Systems at national scale

Records lived in separate systems, operations were entered by hand, and there was no single picture of cash obligations or cash flow — decision-makers could not see the real state of the business at any given moment.

• Automated the capture of operations and streamlined their processing — manual routine taken off the managers.
• Two-way reconciliation at every step: every obligation between counterparties is recorded, and discrepancies close themselves.
• Metal trades and every type of FX operation, cross-conversion included.
• Document flow, movements and every kind of financial accounting inside one system.
• End-to-end analytics and dashboards: leadership sees the real state of the business in real time.
• A family of Telegram bots for analytics, data capture and operational shortcuts.

The team’s routine is streamlined, errors and discrepancies are closed by automatic reconciliation, and decisions are made on real-time data.

Inventory and finance rested on one inventory manager’s manual work; there was no control over shifts or individual stores.

• Automatic inventory accounting: intake, sales, returns, stock on hand.
• Every financial operation is recorded automatically; payroll calculates itself.
• Shift tracking: shifts open automatically across stores, with data capture (cleanliness, store checklist, staff uniform).
• Shift and staff analytics: the best and the worst shifts are visible.
• A Telegram bot records sales and returns and keeps the books.
• Financial calendar: upcoming payments, loans, projections driven by inventory data.

A full-time inventory manager role was removed (a permanent payroll saving), human error is minimal, and stores and shifts are transparent in real time.

A stream of inbound from people who just want to understand the basics first; managers drowning in initial enquiries; no single picture of clients and properties.

• A Telegram bot runs first-line support: explains, answers routine questions, warms leads up.
• Qualified enquiries land in the CRM automatically, where managers take the deal forward.
• The CRM shows only what matters: contact history, statuses, properties and floor plans, the sales plan.
• Automatic document generation.
• Dashboards for owners and decision-makers — the whole sales and property picture in real time.

The bot closes the first line, so managers only work warm leads; the funnel and the properties are fully transparent; paperwork is automated.

Orders lived in chats and on paper, trip statuses were discovered by phone call, trip paperwork took days to assemble, and settlements with customers and hired transport were reconciled by hand, with errors.

• One loop for orders and trips: from the customer’s enquiry to closing the trip, with statuses and history.
• A Telegram bot for drivers: the trip, its stops, documents and photo evidence — statuses update from the road.
• Automatic trip paperwork: orders, contracts, invoices and acts generated from system data.
• Payment and settlement control: who owes the company and whom it owes, with overdue items highlighted.
• Dispatcher and management dashboards: trips in progress, margin by route, fleet utilisation.

A dispatcher runs several times more trips without losing control, trip paperwork is ready the day the trip closes, and settlements reconcile themselves.

Orders arrived by e-mail and messenger, managers checked stock by phoning the warehouse, reservations went missing, and 1C lived apart from sales — constant mis-picks and failed shipments.

• Order intake in one loop: e-mail, messengers and the customer portal converge into a single queue.
• Automatic reservation: an order sees real stock immediately and reserves it — no calls to the warehouse.
• Two-way 1C integration: catalogue, stock, shipments and payments sync themselves.
• A supply loop: shortfalls on orders turn into purchase and production requests.
• A B2B customer portal: the client sees their contract prices, stock, order statuses and documents.

An order is handled in minutes instead of hours, reservations and stock agree, managers sell instead of reconciling, and shipments stopped failing over mis-picks.

Receptionists could not keep up with replies and bookings, clients forgot their appointments, branch utilisation was uneven, and nobody was measuring whether clients came back.

• An AI receptionist on Telegram and WhatsApp: answers, finds a slot, books and reschedules around the clock.
• Smart scheduling by branch and specialist: gaps fill automatically, double bookings are impossible.
• Automatic reminders and confirmations — no-shows drop without anyone making calls.
• Network analytics: branch and specialist utilisation, retention, average ticket — on one dashboard.
• A single client base with visit history, so marketing has something to work with.

Booking runs around the clock without receptionists, no-shows fell noticeably, branch utilisation evened out, and the network is managed on numbers.