Setting Up Automatic Reordering from Suppliers in 1C-Bitrix
Imagine a manager spending 3 hours a day manually monitoring stock levels across 80,000 items. An item runs out, but the replenishment order isn't placed — lost sales and emergency deliveries at your expense. Or the opposite: ordering "by eye" ties up working capital. We solve this through automatic ordering based on CommerceML and native Bitrix mechanisms.
Our architecture eliminates the human factor: an agent checks stock, compares it with the reorder point, and creates an order. The order is sent via API, email, or EDI — depending on the supplier's capabilities. Over years of implementations, we've reduced purchasing time by 70% for retail and distribution clients.
Problems We Solve
A typical scenario: average lead time is 10 days, but the reorder point is not set — by the time the goods arrive, you've already run out. Or the opposite: the reorder point is too high, tying up 40% safety stock instead of 20%. Manual control leads to errors:
- Manually monitoring thousands of items takes up to 3 hours a day.
- Different sending formats (email, API, EDI) for each supplier.
- Duplicate orders due to integration failures.
We address each problem architecturally: predictive reorder points, an agent with a NOT EXISTS safeguard, and flexible sending adapters.
How the Reorder Agent Works
The AutoReorderAgent runs on a schedule (hourly or daily) and executes a query: it selects products whose current stock is below the reorder point and for which there is no active order with the same supplier. The NOT EXISTS condition is key — it prevents duplicates even if the cron job fails.
function AutoReorderAgent(): string { $db = \Bitrix\Main\Application::getConnection(); $result = $db->query(" SELECT rs.product_id, rs.reorder_qty, rs.supplier_id, SUM(sp.AMOUNT) AS current_stock FROM bl_reorder_settings rs LEFT JOIN b_catalog_store_product sp ON sp.PRODUCT_ID = rs.product_id WHERE rs.active = 1 GROUP BY rs.product_id, rs.reorder_qty, rs.supplier_id HAVING current_stock < rs.reorder_point AND NOT EXISTS ( SELECT 1 FROM bl_supplier_orders so WHERE so.product_id = rs.product_id AND so.status IN ('pending', 'sent', 'confirmed') ) "); while ($row = $result->fetch()) { SupplierOrderService::create( $row['supplier_id'], $row['product_id'], $row['reorder_qty'] ); } return 'AutoReorderAgent();'; } Why Use Predictive Reorder Points?
A simple fixed number ignores seasonality and trends. We use average sales over the last 90 days, multiply by lead time, and add a 20% safety stock. The formula is recalculated automatically once a week via an agent. Results: forecast accuracy of 85–90% and a 30% reduction in safety stock.
How We Eliminate Order Duplication
Duplication is a common problem in self-built automation. We implement double protection:
- In the agent's SQL query — the
NOT EXISTScondition (a new order is not created while the previous one is inpending,sent, orconfirmedstatus). - Blocking logic: if an order is not confirmed by the supplier within 24 hours, the agent does not create a duplicate but notifies the administrator.
We also maintain a log of all operations (following official Bitrix agent creation guidelines).
How We Do It: A Real Case Study
Our client — a chain of 15 auto parts stores with a catalog of 80,000 items and 200 suppliers. They needed to automate ordering via API for the top 10 suppliers and via email for the rest. We designed three tables:
-
bl_reorder_settings— thresholds per product linked to supplier. -
bl_supplier_orders— order history with statuses. -
bl_supplier_shipping_methods— method and configuration for sending.
The agent creates a record in bl_supplier_orders, then SupplierOrderService picks the appropriate adapter (API/Email/EDI) and sends it. After the supplier confirms, the status is updated via webhook or manually. When goods arrive, the manager creates a receipt document — stock levels adjust automatically.
Result: managers stopped spending 3 hours a day on ordering; stockout rate dropped from 12% to 2%. The project paid for itself in 3 months due to reduced downtime and emergency shipping costs.
What's Included in the Work
| Component | Description |
|---|---|
| Analysis | Requirements gathering, audit of current stock and suppliers |
| Design | Database schema, agent architecture, and adapters |
| Development | Tables, agent, sending classes, admin interface |
| Testing | Duplicate checks, edge cases (zero stock, API errors) |
| Deployment | Agent schedule setup, migrations, documentation |
| Training | Instructions for managers on order confirmation and receipt |
Estimated time: from 5 to 20 days depending on the number of suppliers and integration methods. Pricing is determined after analysis. Contact us — we'll assess your project within 1 day.
Comparison: Manual vs. Automatic Ordering
| Parameter | Manual Ordering | Automatic Ordering |
|---|---|---|
| Time spent on purchasing per day | 3-4 hours | 15 minutes (oversight) |
| Stockout rate | 12-15% | 2-3% |
| Emergency shipping costs | 5-7% of turnover | 0.5-1% |
| Order error rate | 5-8% | <1% |
Typical Mistakes in Self-Setup
- Missing duplicate-order lock: the agent creates a new order while the previous one is still pending. - Threshold without considering lead time: delays cause stockouts. - Ignoring API errors: the order isn't sent, but its status remains "sent". We add logging and retries.We guarantee stable operation: after commissioning, we provide 30 days of support. Our team has 10+ years of experience with Bitrix. Get a consultation to learn how automation can solve your procurement challenges.