This page explains re-forecasting: comparing a style's own real early-season sales against its original standing forecast, and adjusting the remaining season's production plan once a genuine gap appears. It is for a Make to Stock buyer or planner running a standing forecast that its own factory produces against continuously through a season.
At work, a domestic buyer such as Cedar & Co. sets a standing forecast before a season starts and reorders against it continuously as the season runs. Real sales rarely match a forecast exactly from the first week; the real question is not whether the forecast was perfectly right, but whether the gap between the forecast and real early sales is large enough, and consistent enough, to justify changing the plan for the rest of the season. Holding an original forecast unchanged once real sales clearly show a different pattern is a real problem: it wastes the very early sales data a Make to Stock model is meant to use. This lesson gives you the method for deciding when and how to adjust a standing forecast once real sales data exists.
You need: lessons 3.1 and 3.2's forecasting and phasing method are useful background but not required.
By the end of this lesson you can:
Not covered here: building the original standing forecast from sales history (lesson 3.1); the cost of a forecast that stays wrong in either direction (lesson 3.4).
| Term | Meaning |
|---|---|
| Standing forecast | The demand estimate an MTS buyer, like Cedar & Co., builds its own regular orders against, rather than placing one order at a time. |
| Re-forecast | Updating the remaining part of a season's forecast once real sales data shows a sustained gap from the original estimate. |
| Sales gap | The difference between a style's own real recorded sales and what its standing forecast predicted for the same period. |
| Sustained pattern | A sales gap that continues across more than one recording period, rather than appearing in a single week only. |
| Reorder point | The quantity of stock remaining at which a Make to Stock buyer places its next production order against the standing forecast. |
| Bias | A forecast that is consistently too high or too low in the same direction across several periods, rather than randomly off in either direction. |
A standing forecast is the starting estimate a Make to Stock buyer orders against, but it is not fixed for an entire season. Once real recorded sales show a sales gap from that estimate that continues across more than one period, a genuinely sustained pattern rather than one unusual week, the remaining part of the season's forecast is re-forecast to reflect what is actually happening. A gap in one single week, without a second week confirming it, is ordinary variation, not yet a reason to change the plan.
| What can differ | The shapes you will meet | Why it matters |
|---|---|---|
| Domestic, Make to Stock, continuous | Cedar & Co. checks its own standing forecast against real weekly sales across its 12 shops and re-forecasts the remaining season once a two-week gap is confirmed. | A Make to Stock buyer has the continuous sales data a re-forecast needs, since it sells and restocks throughout the season rather than after one shipment. |
| Export, Make to Order, single order | Northline Apparel's Order A confirms one quantity before production starts, with far less scope to re-forecast mid-order once fabric and cutting are already committed. | A Make to Order order has much less room to change mid-production; re-forecasting there mainly informs the next order, not the current one. |
| A gap in one week only | A single unusually strong week driven by a one-off local event, followed by a return to the original forecast's own pace the next week. | A one-week gap without a second period confirming it is ordinary variation, not a signal to re-forecast the rest of the season. |
| A gap that shows real bias | Real sales run consistently above the standing forecast for three straight weeks, a genuine, sustained pattern rather than noise. | A sustained gap across multiple periods is the specific condition that justifies changing the remaining season's plan. |
Cedar & Co. set a standing forecast of 1,200 pieces for the first four weeks of the season for a basic T-shirt style similar in construction to Order A. After two weeks, real sales across its 12 shops total 760 pieces, ahead of the pace the four-week forecast assumed. Cedar & Co.'s buyer checks whether this two-week result is a sustained pattern or ordinary variation before deciding whether to re-forecast the remaining two weeks of the period.
| Question | Holding the original forecast unchanged | Checking real sales against the forecast |
|---|---|---|
| What does week 3 and 4 production assume? | The original 1,200-piece pace, regardless of what the first two weeks actually showed | Whatever the real two-week result indicates about the remaining weeks |
| What happens if real demand is genuinely running ahead? | Shops likely run short in weeks 3 and 4, since production was never adjusted | The remaining weeks' production is sized to the real, confirmed pace |
| What happens if the fast two weeks do not continue? | No difference, since the plan was never adjusted in the first place | A re-forecast made too early on too little data could overcommit stock |
Cedar & Co.'s buyer compares the real two-week sales rate against the standing forecast's own implied weekly rate to size the sales gap.
A common mistake is treating one strong week as proof the whole season's forecast is wrong and re-forecasting immediately, before a second week confirms whether the pattern actually continues. If Cedar & Co. had re-forecast its remaining two weeks after only week 1's own result, without waiting to see whether week 2 confirmed the same pace, a one-off strong week driven by a local event could have led to an over-committed production order for weeks 3 and 4, with any excess resolved only through a later markdown.
A standing forecast is a starting estimate, not a fixed number a buyer must live with for a whole season regardless of what real sales show. Once real sales are recorded, they should be compared against the forecast's own rate for the same period. If the gap between real sales and the forecast holds across more than one period, it is a genuine, sustained pattern, and the remaining part of the season should be re-forecast using the real, confirmed rate instead of the original assumed one. A gap in a single week alone is ordinary variation and is not yet a reason to change the plan.
Cedar & Co.'s buyer has confirmed real sales running at 380 pieces per week against the original 300-piece forecast rate, holding across both of the first two weeks.
Should Cedar & Co. place the weeks 3 and 4 order at the original 600 pieces, or at the re-forecast 760 pieces? Decide before you open the answer.
At the re-forecast 760 pieces. The 80-piece weekly gap has held across both of the first two weeks, meeting the sustained-pattern condition; this is not a single-week spike being mistaken for a genuine trend.
First, why the original 600-piece figure is now out of date. The 600-piece plan for weeks 3 and 4 assumed the original 300-piece weekly rate, a rate the confirmed two-week real result has already exceeded consistently, not just once.
Second, why re-forecasting now, rather than waiting, matters. Cedar & Co.'s production slot for weeks 3 and 4 is still open; re-forecasting before it closes lets the order reflect the confirmed real rate, while waiting risks placing the outdated 600-piece order and running short mid-period.
Cedar & Co. should place the weeks 3 and 4 order at 760 pieces, the re-forecast figure built from the confirmed real rate, and continue checking real sales against this new rate as the remaining weeks unfold. Lesson 3.4 covers what a forecast that stays wrong in either direction actually costs.
Framed in lost sales rather than a single penalty figure: the gap between the outdated order and the re-forecast figure is the stock Cedar & Co. would be missing if real demand continues at its confirmed rate.
A 160-piece shortfall against confirmed real demand, left unaddressed because the original forecast was never checked against real sales, converts directly into approximately USD 3,184 of lost sales across weeks 3 and 4 alone, without counting any shops running short of the style for customers who might otherwise have returned later in the season.
The workbook's "Re-forecast" sheet takes the original standing forecast's own weekly rate, compares it against real sales entered period by period, flags whether the gap has held for more than one period, and calculates a re-forecast quantity for the remaining weeks once the sustained-pattern condition is met.
Cedar & Co.'s own rows record the 300-piece forecast rate against 380-piece real weeks 1 and 2, with the sheet flagging a sustained gap and calculating the 760-piece re-forecast for weeks 3 and 4.
1. Why is a standing forecast re-forecast only once a sales gap holds across more than one period, not after a single strong or weak week?
2. Cedar & Co.'s style sold 760 pieces in the first two weeks against a 1,200-piece, four-week forecast. What is the real weekly sales gap?
3. After confirming an 80-piece weekly sales gap across two straight weeks, what should Cedar & Co. do for the weeks 3 and 4 production order?
| EN | AR | FR | TR | Meaning |
|---|---|---|---|---|
| Standing forecast | توقعات ثابتة | Prévision permanente | Sürekli tahmin | The demand estimate an MTS buyer, like Cedar & Co., builds its own regular orders against, rather than placing one order at a time. |
| Re-forecast | إعادة التوقع | Reprévision | Yeniden tahmin | Updating the remaining part of a season's forecast once real sales data shows a sustained gap from the original estimate. |
| Sales gap | فجوة المبيعات | Écart de ventes | Satış farkı | The difference between a style's own real recorded sales and what its standing forecast predicted for the same period. |
| Sustained pattern | نمط مستمر | Schéma soutenu | Sürekli örüntü | A sales gap that continues across more than one recording period, rather than appearing in a single week only. |
| Reorder point | نقطة إعادة الطلب | Point de recommande | Yeniden sipariş noktası | The quantity of stock remaining at which a Make to Stock buyer places its next production order against the standing forecast. |
| Bias | الانحياز | Biais | Sapma (yanlılık) | A forecast that is consistently too high or too low in the same direction across several periods, rather than randomly off in either direction. |
You now know how to adjust a forecast once real sales data starts coming in. The next lesson quantifies what a forecast that is wrong in either direction, too high or too low, actually costs. Back to the course map