What is Suggested Order Quantity (SOQ)?

Suggested order quantity (SOQ) is the quantity of a product a system recommends ordering, calculated from what is on hand, what is already on its way, how fast the item sells, how long resupply takes, and the stocking rules the business has set. It converts ordering from a memory exercise into a proposal that can be reviewed, adjusted and explained.

Why suggested order quantity matters

Ordering by feel produces two expensive errors at once. Fast-moving lines run out — costing the sale and, in retail, often the shopper — while slow-moving lines accumulate, tying up working capital in stock that will not turn. Both errors can be present in the same outlet on the same day.

A suggested quantity attacks the root cause: the person ordering rarely has the numbers in front of them. A rep standing in a shop cannot recall what that outlet sold over eight weeks, what is in transit, or what the last three orders looked like. A system can.

How suggested order quantity is calculated

Most implementations are a variation on target cover minus current position:

Demand rate — average consumption per day or week, usually over a trailing window long enough to smooth noise but short enough to track trend.
Target cover — how many days of demand the location should hold, which is a policy decision, not a calculation.
Lead time — how long resupply takes, because the order must cover demand until the next delivery lands, not just until tomorrow.
Current position — stock on hand, plus everything already committed: in-transit shipments, open purchase orders, unfulfilled requisitions.
Rounding rules — minimum order quantity, case or lot size, and any maximum.

A workable form is: suggested quantity = (demand rate × (target cover + lead time)) − stock on hand − incoming − open commitments, rounded up to the lot size and clipped to the minimum and maximum. The subtraction of incoming stock is the step most often omitted, and omitting it is what causes a system to recommend re-ordering something that is already on a truck.

An example

An outlet sells 12 units a week of an SKU. Target cover is two weeks, lead time is one week, so the target position is 12 × 3 = 36 units. There are 9 on hand and 6 already on an undelivered order, giving a current position of 15. The suggested quantity is 36 − 15 = 21, rounded up to a case size of 24. Had the 6 in transit been ignored, the suggestion would have been 27 — a 25% over-order on a single line, repeated across a catalogue.

Common variations

Min/max replenishment. Order up to a maximum whenever stock falls below a minimum. Simple, robust, and insensitive to trend.
Reorder point driven. Trigger a fixed quantity when stock crosses the reorder point.
Consumption-based forecasting. Project forward from observed movement, optionally weighted to recent periods.
Statistical or machine-learning forecasting. Model demand from historical sales, movement patterns, seasonal variation and other influencing parameters, rather than projecting a trailing average. Worth the extra machinery only where demand has a shape to learn.
Constrained suggestion. Cap the total by credit limit, budget, or vehicle capacity — common in van sales, where the van is the constraint.

Limitations worth stating

A suggested quantity is only as good as its inputs. If average consumption and lead time are maintained by hand and never revisited, the arithmetic will be confident and wrong. Promotions break trailing averages, because past demand stops predicting future demand precisely when volumes matter most. And new items have no history at all, so they need a seeded assumption rather than a calculation. A good implementation therefore shows its working, so the person accepting the number can see which assumption to distrust.

How xMatix supports suggested order quantity

xMatix Procurement generates suggested order lines through three replenishment engines — minimum stock level, replenish-to-maximum, and consumption-based forecasting — plus template orders for repeating patterns. Suggestions can be produced on a schedule, so a dated run exists to review rather than being triggered manually.

The calculation nets the full pipeline: stock on hand, in-transit shipments, pending purchase orders, open requisitions and unfulfilled sales orders. Items can be included or excluded by ABC and FSN classification, so fast, slow and non-moving lines can be treated differently in the same run.

Every suggested line carries its own explanation — projected quantity, quantity available, each pipeline component, lead time and average consumption are written onto the line itself. The number can always be traced back to the arithmetic that produced it, which is what makes it safe to accept in bulk. Separately, when a purchase order line would order stock the pipeline already covers, Sense Assist raises a nudge that states the netting and offers the corrected quantity in one click.

Stocking parameters — minimum and maximum levels, reorder level and quantity, minimum and maximum order quantity, lot size, lead time and average consumption — are held per item, so the same engine can behave differently by product without custom code.