When to Reorder a Bestseller: What the Reorder Point Formula Decides
A hoodie is selling 250 units a week, there are 12 weeks of cover on the shelf and another 1,000 units are in transit, and the Monday meeting needs a yes or no on the repeat. Order now and you may pay for stock the season cannot absorb. Order later and the style can run out while the repeat is still on the water. The reorder point formula turns that argument into a number: the level of stock, counting everything already on order, at which a repeat has to be placed so that it lands before the shelf is empty.
The discipline is in the inputs: a repeat takes the full replenishment lead time to arrive, so you order while stock can survive that wait, plus a buffer for faster sales or later deliveries.
Those tiles come from the worked example later in this guide and are illustrative, not benchmarks. The structure matters more than the values.
The four steps of a reorder decision
- 1Measure velocity and coverNet weekly sales from in-stock weeks, then weeks of cover on hand and on order.
- 2Add up the lead timeDecision to goods available to sell, including every wait in between.
- 3Size the safety stockUse demand variability and lead-time variability, not a flat rule of thumb.
- 4Run two scenariosTest a stronger and a weaker demand path against the season end.
A framework for the decision, not a fixed process. Programmes with fabric held in advance or air freight options will weight the steps differently.
The Reorder Point Formula, Variable by Variable
In its standard inventory form the reorder point is the demand you expect during the replenishment lead time plus safety stock. It is compared with the inventory position, which is stock on hand plus stock on order and in transit, minus anything already committed to orders or allocations that will not be available to sell. When the position falls to the reorder point, you place the order.
Reorder point = (average weekly sales x lead time in weeks) + safety stock. Inventory position = on hand + on order + in transit. Place the order when the inventory position is at or below the reorder point.
Comparing the reorder point with on-hand stock alone is the most common error, because it ignores units already coming. The table defines each variable and where it usually goes wrong.
Reorder point formula: inputs and how they go wrong
| Input | Definition | Common error |
|---|---|---|
| Average weekly sales (velocity) | Net units sold per week over a stated window, from weeks when the style was in stock | Averaging weeks that include stockouts, which understates demand |
| Replenishment lead time | Weeks from the decision to order until the goods are available to sell | Quoting production time only and forgetting approval, transit and receiving |
| Safety stock | Buffer for demand and lead-time variability, set for a chosen service level | A flat "two weeks" rule that ignores how volatile the style is |
| Inventory position | On hand plus on order plus in transit, less committed or reserved units | Counting only the stock in the warehouse |
Definitions follow standard inventory practice. Retailers differ on how they treat reserved and returned units, so confirm your own definitions before building a template.
Step 1: Calculate Weekly Sales Velocity and Weeks of Cover
Sales velocity is units sold per week, but the figure you need is net demand: sales adjusted for returns and for weeks when stock was short. A style that sold out on Thursday shows low sales for a week with unmet demand behind them. Use in-stock weeks where you can, and state the window, because the choice moves the answer.
Illustrative eight-week sales history for one hoodie colour
| Measure | Value | How it is calculated |
|---|---|---|
| Weekly units, weeks 1 to 8 | 190, 310, 240, 200, 280, 230, 300, 250 | Illustrative net sales, all weeks in stock |
| Eight-week average | 250 units | Sum of 2,000 divided by 8 weeks |
| Weekly standard deviation | About 44 units | Sample standard deviation of the eight weeks, rounded |
| Effect of window choice | 120 more units over the lead time | A last-four-weeks average of 265 (1,060 / 4) against 250, times 8 weeks |
Illustrative data built for this example. Short windows react faster to a trend but are noisier; long windows are steadier but lag a fading or accelerating style.
Weeks of cover is available stock divided by forecast weekly sales. State which stock you mean. Cover on hand is 3,000 divided by 250, which is 12 weeks. Forward cover, which also counts the 1,000 units in transit, is 4,000 divided by 250, which is 16 weeks. The decision uses forward cover, because the units in transit will be there before the new order lands. Check how your trading report defines it; our guide to reading Monday trading reports covers how those columns are built.
Adjust for known changes such as a promotion, a price move or a seasonal turn: history tells you what happened, and the reorder point needs what will happen over the next lead time.
Step 2: Pipeline Stock and Total Replenishment Lead Time
Pipeline stock is everything ordered but not yet available to sell: units in production, in transit and in customs or goods-in. It counts towards the inventory position only if you trust its arrival date, so ask for the confirmed ex-factory date and the booked sailing or flight, not the original plan.
Replenishment lead time starts when you decide, not when the purchase order is signed, and ends when the goods can be sold, not when they leave the factory. The timeline splits a hypothetical eight weeks into its usual stages; each programme must re-time it. Our clothing production lead times guide explains what drives each stage.
Illustrative 8-week replenishment lead time
- Week 1: decision and confirmationReorder approved, quantities by colour and size agreed, purchase order confirmed, any approvals needed on the existing bulk standard.
- Weeks 2 to 4: fabric and trimsFabric and trims are booked or released from reserved stock. If the yarn or fabric must be made again, this block is where lead time grows.
- Weeks 5 to 6: cut, sew, finish and packProduction on the confirmed standard, with inline and final inspection before release. See how to manufacture clothing in bulk for the stages.
- Weeks 7 to 8: transit and receivingBooking, shipment, customs, delivery to the warehouse and goods-in, so the stock is available to sell.
Hypothetical split that sums to the 8 weeks used in this guide. Do not use it as a typical lead time; ask each factory and forwarder for current, confirmed timings.
Shortening the lead time lowers the reorder point directly, because less demand falls inside it. That is why buyers reserve fabric or capacity for proven winners, at the price of commitment, so reserve only what the sales data supports.
Step 3: Set Safety Stock Based on Demand Variability
Safety stock covers the gap between forecast and reality during the lead time. The standard calculation multiplies a service-level factor, called z, by the standard deviation of demand over the lead time. For weekly demand with a standard deviation of sigma, and a lead time of L weeks, the standard deviation over the lead time is sigma multiplied by the square root of L. Using the example above, 44 multiplied by the square root of 8 is about 124 units, and a z value of 1.645 gives 205 units of safety stock.
The z value is a standard normal-distribution factor: 1.28 corresponds to a 90 per cent chance of not stocking out during the lead time, 1.645 to 95 per cent and 2.33 to 99 per cent. These are conventions, not recommendations. Choose the level from the cost of a stockout against the cost of carrying stock, higher for proven core styles than for uncertain newness.
Illustrative safety stock by service level (8-week lead time)
| Service level | z value | Demand variability only (units) | With 1-week lead-time variability (units) |
|---|---|---|---|
| About 90 per cent | 1.28 | 159 | 357 |
| About 95 per cent | 1.645 | 205 | 459 |
| About 99 per cent | 2.33 | 290 | 651 |
Illustrative: weekly demand of 250 units, standard deviation of 44 units, lead time of 8 weeks, lead-time standard deviation of 1 week. The combined formula is z multiplied by the square root of (L x sigma squared + weekly sales squared x lead-time sigma squared).
Read the last column twice. Allowing for a lead-time standard deviation of just one week more than doubles the buffer in this example, which is why late deliveries hurt more than a noisy sales line. If your factory's on-time record is patchy, the cheapest safety stock is often a better delivery plan, not more units; see test and scale for how capacity can be agreed before a winner is known.
Worked Example: The Reorder Point Formula for a Hoodie Selling 250 a Week
Here is the reorder point formula applied to one hoodie colour. All inputs are illustrative assumptions, set out in the table, and the arithmetic is shown in full so you can swap in your own numbers.
Illustrative inputs and the reorder point calculation
| Line | Value | Working |
|---|---|---|
| Average weekly sales | 250 units | Eight-week average from Step 1 |
| Replenishment lead time | 8 weeks | Decision to available to sell (assumed) |
| On hand today | 3,000 units | Assumed |
| In transit, arriving week 4 | 1,000 units | Assumed, confirmed booking |
| Inventory position | 4,000 units | 3,000 + 1,000 |
| Demand during lead time | 2,000 units | 250 x 8 |
| Safety stock (about 95 per cent) | 205 units | 1.645 x 44 x square root of 8 |
| Reorder point | 2,205 units | 2,000 + 205 |
| Reorder point in weeks | 8.82 weeks | 2,205 / 250 |
| Season end | Week 20 | Assumed, counted from today |
All figures illustrative and rounded; the example does not represent any real style, sales rate, supplier lead time or result.
The position of 4,000 units is above the reorder point of 2,205, so the answer today is not yet. The position falls by 250 units each week, so it reaches the reorder point after about 7.2 weeks. Of the 2,205 units, 2,000 (90.7 per cent) is plain demand during the wait and only 205 (9.3 per cent) is buffer, so the lead time input matters far more than the safety stock.
Weekly Review: Why the Trigger Should Sit Above the Reorder Point
A buying team reviews the numbers weekly, so if the position crosses the reorder point on Tuesday, nothing happens until the next review. The fix is to raise the trigger by one review period of demand, so the order goes in at the last review before the position would drop below the reorder point.
Illustrative weekly reviews for the base plan
| Week | On hand | Inventory position | Weeks of forward cover | Action |
|---|---|---|---|---|
| 4 | 3,000 | 3,000 | 12.0 | Hold |
| 5 | 2,750 | 2,750 | 11.0 | Hold |
| 6 | 2,500 | 2,500 | 10.0 | Hold |
| 7 | 2,250 | 2,250 | 9.0 | Order now |
| 8 | 2,000 | 2,000 | 8.0 | Past the trigger |
Illustrative. The 1,000 units in transit arrive in week 4, so on hand rises while the position does not. Trigger = reorder point of 2,205 plus one week of sales (250) = 2,455 units.
In week 7 the position is 2,250, which is at or below the 2,455 trigger, so the order goes in. It lands in week 15, when the shelf holds 250 units, about one week of sales. Had the team waited for the strict reorder point of 2,205, the next review is week 8 with a position of 2,000, and the order would arrive in week 16 with the shelf empty. The 205-unit safety stock was consumed by the review gap alone.
Step 4: Model Optimistic and Conservative Scenarios Against the Season End
One forecast is a single bet. Run the same formula against a stronger and a weaker demand path, and read each against the season end, because a repeat that lands after the selling window is stock without a selling season. The latest useful order week is the season end minus the lead time, here week 12.
Illustrative scenarios for the same hoodie
| Measure | Optimistic: demand rises | Base | Conservative: demand fades |
|---|---|---|---|
| Weekly sales | 300 | 250 | 180 |
| Demand during 8-week lead time | 2,400 | 2,000 | 1,440 |
| Reorder point (safety stock held at 205) | 2,605 | 2,205 | 1,645 |
| Trigger (reorder point plus one week) | 2,905 | 2,455 | 1,825 |
| Order placed at week | 4 | 7 | 13: too late |
| Lands at week | 12 | 15 | 21: after season end |
| Order quantity to reach season end | 2,000 | 1,000 | No reorder |
| Units left at week 20 | 0 | 0 | 400 (markdown exposure) |
Illustrative. Season end is week 20. Order quantity = weekly sales x weeks from arrival to season end, less stock on hand at arrival. Safety stock is held constant for comparison; recalculate it for each demand path in real use.
The scenarios say something a single forecast cannot. If demand rises, the order has to go in at week 4 and be twice the base quantity; if demand fades, the prudent response is no order at all. The chart shows each trigger week against the last useful order week, and that gap is the margin for judgement.
Week the weekly-review trigger is reached (illustrative)
Weeks from today
Illustrative scenarios from the table above, not a forecast. A scenario whose trigger week is later than the latest useful order week should not trigger a reorder in this season.
There is a second stress test: a fade that starts after the order is committed. If the base order of 1,000 units is placed in week 7 and demand then drops to 180 a week, the season ends with 910 units still in stock. That is the exposure to price before approving a repeat, and the reason for staged repeats on uncertain styles; our guide to reducing overproduction without reducing sales goes further.
The Cost of Ordering Too Late or Too Early
A late order appears as a stockout and lost sales in the trading report. An early order shows up weeks later as slow cover, markdown pressure and cash tied in stock that cannot yet sell. Compare both in the same units before choosing a trigger.
Illustrative cost of timing errors for the base plan
| Timing | What happens in this example | Where the cost shows up |
|---|---|---|
| Reorder at the weekly trigger (week 7) | Lands in week 15 with 250 units on the shelf; no stockout | Little visible cost if lead time holds |
| Reorder when forward cover reaches zero (week 16) | Lands in week 24, after the week-20 season end; weeks 16 to 20 are lost | Lost sales of 1,000 units in-season (4 weeks x 250) |
| Reorder at the trigger, then a 3-week delay | Lands in week 18; shelf empty from week 16 | Lost sales of 500 units (3 weeks x 250, less 250 on the shelf) |
Illustrative arithmetic on the assumed base plan. Assumes lost sales are not recovered later and customers do not switch to another style; real behaviour differs.
To turn lost units into money, multiply by gross profit per unit, which is selling price excluding VAT minus cost. At an assumed GBP 25 per hoodie, 500 lost units is GBP 12,500 of gross profit. Gross margin is that profit divided by the selling price excluding VAT; markup divides it by cost instead, so do not mix the two.
Reorder Point Formula KPIs: Stock Cover, Lost Sales and Reorder Lead Time
Three measures keep the formula honest after the first decision. Define each before the meeting, because the same word can mean different things in different systems, and state the formula in the template so everyone reads the same number.
KPIs to monitor on every repeat decision
| KPI | Definition and formula | What it tells you | Watch for |
|---|---|---|---|
| Stock cover (weeks of cover) | Available stock divided by forecast weekly sales; state whether it is on hand or includes stock on order | How long the shelf lasts against the lead time | Cover at or below lead time plus review period means a repeat is due now |
| Lost sales | Estimated demand not served during stockout weeks: forecast weekly sales x weeks out of stock, adjusted for any units backordered | The cost of ordering late, in units and in gross profit | It is an estimate; compare with in-stock rate and customer comments |
| Reorder lead time | Weeks from the order decision to the goods being available to sell, measured on actual orders | Whether the lead time you plugged into the formula is real | A rising average or a wide spread means safety stock is too low |
Definitions follow common retail planning practice but vary by retailer. Confirm whether your cover figure uses on-hand stock, available-to-sell stock or includes the pipeline.
Compare planned and actual reorder lead time after every repeat. If actuals drift upwards, the formula is still right but the input is stale, and the reorder point must move before the next stockout.
Decision Rules: What to Do and What Evidence Would Change It
Turn the calculation into a rule your trading team can apply in a few minutes, and write down what would make you override it.
A working reorder decision checklist
- Velocity is net demand from in-stock weeks, with the window stated and a forecast adjustment for any known change.
- Position includes confirmed pipeline only, with ex-factory and arrival dates checked this week.
- Lead time runs from decision to available to sell, and is checked against the last few actual repeats.
- Safety stock is set per style for a stated service level, with lead-time variability included.
- Trigger is the reorder point plus one review period of sales.
- Both scenarios are run against the latest useful order week, and the residual stock is priced.
- Fabric, trims and capacity for the repeat are confirmed with the factory before the order is released.
A starting template. Add your own rules on returns, size curve and colour split before it goes into a trading pack.
Evidence that should change the decision: a rise in returns on the style, a size curve selling unevenly so that some sizes stock out while others sit, or a lead time the factory cannot confirm. Check the minimum quantity before sizing the order, since our guide to MOQ in clothing manufacturing shows how minimums can force a repeat larger than the formula suggests.
For a hoodie, fleece lead times and seasonality deserve their own treatment, including on-order units and season-end markdown risk. Our hoodie reorder calculation goes deeper there, while this page covers the general method. To decide whether a style is a winner at all before you apply the formula, start with which styles deserve a bigger buy.
A final question for your next trading meeting: if the style you are watching sold at your conservative rate from today, would the repeat you are about to approve still land inside the selling window, and what would you do with the units left over?
How Ruhrose Approaches Repeat Orders
Ruhrose develops hoodies, jersey, denim and woven casualwear for UK buyers, with development and production in Bangladesh. We sample before bulk, and on a repeat we start from the same questions as this guide: the lead time from your decision to goods available to sell, the fabric and trim position, and the capacity available for a winner. We do not quote lead times or minimums in guides because they depend on the fabric, the colour and the programme, so we confirm them for your specific style. If you are testing a style first, the test and scale model sets out a risk-reduction approach, not a risk-free one, and a repeat plan can be agreed alongside the first order.