Inventory Reorder Point Automation
REPLENISHMENT DECISION
Automate a reviewed rule, not a guessed threshold
A reorder point should describe demand during a realistic replenishment lead time plus an agreed uncertainty buffer. The rule also needs a trusted availability calculation, order multiple, supplier path and exception owner.
01 – DEMAND
Choose the demand signal
Use a representative period and adjust for stockouts, seasonality, promotions and known future orders.
02 – SUPPLY
Measure real lead time
Track approval, supplier processing, transit, receiving and release, not only the promised shipping time.
03 – CONTROL
Review generated supply
Define who handles missing vendors, excessive quantities, late orders, pack multiples and unusual forecasts.
Pilot twenty items with different behaviours
Include fast, slow, seasonal, new, minimum-order, long-lead and manufactured items, then compare every suggestion with the underlying calculation.
Inventory reorder-point automation should create or recommend supply when projected availability falls below a controlled threshold. A useful starting formula is demand during replenishment lead time plus a safety buffer. That simplified formula is not a universal answer: the demand period, lead-time definition, reservations, inbound supply, minimum order quantity and pack multiple must match the real operation.
Start with the supplier lead-time guide and the inventory KPI scorecard. Use the inventory systems service when orders, stock and purchasing are split across systems.
1. Define projected availability
Do not trigger only from physical on-hand if some units are reserved, quarantined, already promised or in transit. Define the quantity the rule sees. A practical available or forecast position may start with on-hand, subtract committed demand and unusable stock, and add reliable inbound supply due within the planning horizon. Document which order statuses count and when a late purchase stops being treated as available.
| Input | Definition to fix | Quality check |
|---|---|---|
| On-hand | Stock in approved sellable or usable locations | Reconcile to recent count |
| Demand | Sales, production or transfer demand included | Remove cancelled and duplicate demand |
| Inbound | Confirmed supply and expected date | Flag overdue and partial orders |
| Lead time | Elapsed time from decision to usable receipt | Measure actual distribution |
| Buffer | Quantity held for agreed uncertainty | Review service and excess trade-off |
| Multiple | Pack, case, pallet or production batch | Verify base-unit conversion |
2. Calculate a transparent starting point
For a stable item, estimate average daily demand over a representative period and multiply it by the end-to-end lead time in days. Add a safety buffer chosen for the business risk. Keep the calculation visible. A more advanced statistical model may be justified when demand and lead time are volatile, but complexity does not compensate for missing transactions or a supplier lead time copied from a price list.
Example: an item averages four units per working day and the measured end-to-end lead time is ten working days. Demand during lead time is forty units. If the business approves a twelve-unit buffer, the starting reorder point is fifty-two units. If the supplier sells cases of twelve, the proposed order must also respect that multiple and the desired target level. Test the result against actual demand history; do not treat the example as a recommended buffer.
3. Choose min-max, fixed order or demand-linked replenishment
| Rule | Best fit | Watch for |
|---|---|---|
| Min-max | Repeat items with a useful operating range | Excess when max ignores seasonality or pack size |
| Fixed order quantity | Stable economic or supplier batch | Too many or too few cycles as demand changes |
| Replenish to order | Items not normally held or customer-specific supply | Customer timing and reservation control |
| Manufacture to replenish | Controlled BoM and capacity path | Component shortage and batch constraints |
| Manual review | New, intermittent, obsolete or strategic items | Owner delay and undocumented judgement |
Odoo documents reordering rules with minimum and maximum forecast quantities and routes that can create a request for quotation or manufacturing order. It also requires the product, vendor or bill of materials to be configured for the selected route. Those dependencies should appear in the Mitrend test even when another platform is used.
4. Measure lead time end to end
Start the clock when the need can first be acted on, not when the supplier receives the order. Include internal review, order approval, supplier processing, transit, customs where relevant, receiving, quality inspection and system release. Report median or typical performance alongside late cases. A single optimistic lead time will systematically trigger too late.
5. Protect against distorted demand
- Correct periods when stockouts suppressed recorded sales.
- Separate one-off projects, promotions and launch demand from normal run rate.
- Use forward orders and production plans when they are more informative than recent sales.
- Flag new items with insufficient history for manual review.
- Treat seasonal items with comparable seasons or an approved seasonal profile.
- Stop replenishment for discontinued items only after open demand and obligations are checked.
6. Add purchasing constraints
Convert the suggested quantity into a valid purchase or production quantity. Apply supplier minimum order quantity, pack multiple, price break, warehouse capacity and approved order calendar. Do not let the system silently round a small need into a large cash commitment. Show the raw need, adjusted quantity and reason for every adjustment.
7. Test normal and exception paths
- Forecast falls just above, exactly at and below the trigger.
- Confirmed inbound supply is on time, late, partial and cancelled.
- Demand includes a large order, return, promotion and stockout period.
- Vendor or BoM is missing, inactive or changed.
- Order multiple pushes supply above the maximum.
- The item is quarantined, discontinued or location-restricted.
- A user overrides, postpones or rejects the suggestion with a reason.
8. Monitor outcomes, not only generated orders
Review priority stockouts, emergency purchases, excess after replenishment, overdue inbound orders, manual overrides and suggestions rejected. Compare actual demand and actual lead time with the rule assumptions. Shopify exposes sell-through and days-of-inventory measures that can help review outcomes, provided the period and formula are understood. Change a rule through an approval log rather than editing thresholds until the dashboard looks better.
Worked pilot
A distributor selects twenty items across fast, seasonal, slow and long-lead groups. It calculates the rule card, runs suggestions without creating orders for two cycles, and records what a buyer would change. Three items show false availability from overdue purchase orders and two have incorrect case conversions. Those data defects are fixed before automatic purchase creation is enabled. Automation begins only for the items whose suggestion and exception path are explainable.
Sources checked
- Odoo: Reordering rules
- Shopify Help Center: Product analytics overview
- Xero South Africa: Inventory management
Reviewed by
Mitrend Digital editorial team
2026-07-17
Evidence used for this page
Reviewed against current Odoo reordering-rule documentation, Shopify inventory analytics definitions and Mitrend supplier-lead-time guidance. Includes an original rule card, test matrix and monitoring model.
Turn the guide into a practical next step
This resource provides general implementation guidance. Verify platform settings, tax, legal, payment and operational requirements against the current business context before making a live change.