Reference sell-through curve: season-end projection and gap
Takes the curve of past seasons (what share of the final sell-through was reached by each week), divides the current sell-through by it and gets the season-end projection instead of a straight-line pace. Compares the projection with the plan: the gap in percentage points and in pieces of under-sale, a signal “ahead”, “on track” or “behind” and the last week when a decision can still work. It complements “Sell-through” (pace and stock projection) and “Control week”: here the shape of the season comes from your own history.
At a glance
- What question does it answer?
- Where sell-through will end if the season shape repeats past ones, and until when the price can still be changed.
- What data is needed?
- The reference: cumulative share of the final by week since reaching the shelf.
- How to read the result?
- Look at the projected final and the signal: “behind” means an action plan is needed before the last decision week. Example on the default values: Reference by the current week: 40%; Projected final sell-through: 75%; Sell-through needed by this week: 34%; Gap to the plan: +10%. The figures are illustrative: enter your own to see your result.
Why and how to reason
A straight-line continuation of the pace misleads: the season follows a curve whose first weeks are slower. Past seasons give the shape: what share of the final sell-through is reached by each week. Dividing the current sell-through by that share gives the projected final. In the example by week 5 the reference reaches 40%, sell-through is 30%, so the final is 75% against the plan of 85%: a gap of 10 pp, that is 500 pcs of under-sale out of 5,000.
When to use it
Apply weekly in mid-season for each category and channel when you must decide whether to change the price, move stock or reorder. It does not replace “Sell-through” and “Control week”: those count the pace, here the season shape comes from history.
How to read the result
Look at the projected final and the signal, then at the last decision week: after it little of the sales is left on the curve and the price changes little. The first two weeks are noisy.
Common mistakes
A reference of another season or category, weeks from warehouse delivery, a curve not cleaned of stock-outs, different sell-through bases in the reference and in the current figure.
What data is needed?
- The reference: cumulative share of the final by week since reaching the shelf.
- The current week, sell-through, received and plan.
Inputs
| Field | Unit | Value in the example | Meaning |
|---|---|---|---|
| Current week since the goods reached the shelf | weeks | 5 | Count from the day the goods reached the shelf or the site, not from the warehouse delivery date: otherwise the curve shifts. The example value is illustrative. |
| Sell-through now | % | 30 | Sold ÷ received cumulative over the same weeks as in the curve. The base must be the same as in your reference. The example value is illustrative. |
| Received | pcs | 5,000 | Pieces received for sale in the group the curve is built for. The example value is illustrative. |
| Season-end sell-through plan | % | 85 | The planned share sold by the end of the season. Above 100% is impossible. The example value is illustrative. |
| Sell-through base | Sold ÷ received | Pick the same base the curve was built on: figures of different bases are not comparable. The value is only labeled in the result. | |
| Tolerance “on track” | pp | 5 | The deviation of the projection from the plan in percentage points inside which the signal is “on track”. The example value is illustrative. |
| Minimum share of the season ahead for a decision | % | 25 | A price decision can still work if at least this share of the final sales is left on the curve. You set it. The example value is illustrative. |
How it is calculated
| Step | Formula |
|---|---|
| Reference by the week | C(5) = 40 % (linear between table points) |
| Projected final | 30 % ÷ 40 % = 75 % |
| Gap | 85 % − 75 % = 10 % p.p.; × 5,000 pcs = 500 pcs |
| Signal | projection − plan = −10 %; tolerance ±5 % p.p. |
| Decision week | the last week by which the reference is not above 75 % |
Worked numbers
Default values from the app; in the interactive version you enter your own.
- Reference by the current week
- 40 %The share of the final sell-through reached by this week on the curve.
- Projected final sell-through
- 75 %Sell-through now ÷ the reference by this week.
- Sell-through needed by this week
- 34 %Reference × plan: where you must be on the plan.
- Gap to the plan
- +10 %Plan − projected final, p.p. A minus means a reserve.
- Under-sale by the end
- +500Gap × received, pcs.
- Signal
- behind the plan
- Last week for a decision
- week 9
- Sell-through base
- sold ÷ received
- Reference × plan
- Projection at the current pace
Data as a table
| Measure | Reference × plan | Projection at the current pace |
|---|---|---|
| 1 | 5.1 | 4.5 |
| 2 | 11.1 | 9.8 |
| 3 | 17.9 | 15.8 |
| 4 | 25.5 | 22.5 |
| 5 | 34 | 30 |
| 6 | 42.5 | 37.5 |
| 7 | 50.2 | 44.3 |
| 8 | 57 | 50.3 |
| 9 | 63.8 | 56.3 |
| 10 | 69.7 | 61.5 |
| 11 | 75.7 | 66.8 |
| 12 | 85 | 75 |
Trajectory by week
| Week | Reference | On the plan | Projection |
|---|---|---|---|
| 1 | 6 % | 5.1 % | 4.5 % |
| 2 | 13 % | 11.1 % | 9.8 % |
| 3 | 21 % | 17.9 % | 15.8 % |
| 4 | 30 % | 25.5 % | 22.5 % |
| 5 | 40 % | 34 % | 30 % |
| 6 | 50 % | 42.5 % | 37.5 % |
| 7 | 59 % | 50.2 % | 44.3 % |
| 8 | 67 % | 57 % | 50.3 % |
| 9 | 75 % | 63.8 % | 56.3 % |
| 10 | 82 % | 69.7 % | 61.5 % |
| 11 | 89 % | 75.7 % | 66.8 % |
| 12 | 100 % | 85 % | 75 % |
A gap of 10 % p.p. below the plan is 500 pcs of under-sale at 5,000 pcs received. Decide on price or stock transfers before week 9: after it less than 25 % of the final sales is left on the curve.
Sell-through base: sold ÷ received. There is no common norm: build the curve separately by category and channel (stores, site, marketplaces) from your own past seasons, cleaned of stock-outs and promotions.
A typical mistake: taking the curve of another season or category, counting weeks from the warehouse delivery and not from reaching the shelf, and not cleaning past seasons of stock-outs.
Nuances and limits
- A reference of another category or season.
- Weeks counted from delivery, not from the shelf.
- The curve is not cleaned of stock-outs.