A display can look exactly like the brief and still fail to sell a single extra unit; a plain table of folded stock can outsell a carefully built window if the product itself is what customers actually want. The only way to know which display is working is to measure the sales it produces, compared with what would likely have sold anyway.
This lesson gives you the method a shop uses to isolate a display's own effect on sales from everything else that also changes sales at the same time. You will apply it to Delta Garments' own factory outlet store, measuring a two-week display of Order A's own second-quality T-shirt pieces, because a factory outlet's overrun stock is exactly the kind of product whose display effect is easy to measure against a clear baseline.
You need: the idea of a display brief and its non-negotiable elements, from Lesson 4.1, is useful background but not required.
By the end of this lesson you can:
Not covered here: deciding whether to extend, adjust or remove a display based on its results (Lesson 4.3), or comparing two competing display ideas directly against each other (Lesson 4.4). This lesson only teaches how to measure one display's own result.
| Term | Meaning |
|---|---|
| Baseline sales | The sales a product would likely have made without the display, usually measured from a period just before it went up. |
| Sell-through rate | The share of stock placed for a display that has actually sold by a given date. |
| Display lift | The extra sales a display produces above the baseline, credited to the display itself. |
| Attribution risk | The chance that a sales change is wrongly credited to a display when another cause, such as a price change, actually produced it. |
| Display period | The fixed number of days or weeks a display stays up before it is measured or changed. |
| Confounding factor | Any other change happening at the same time as a display that could also explain a change in sales. |
Sales during a display period are never proof of the display's own effect by themselves, because some of those sales would have happened regardless. The display's own real contribution is the difference between the display period's sales and the baseline period's sales, adjusted for any other known change, such as a price cut or a new season starting. A shop that reports a display's total sales as its "result," without subtracting the baseline, always overstates what the display itself achieved.
| What can differ | The shapes you meet | Why it matters to you |
|---|---|---|
| Business model | A Make to Order export brand like Northline Apparel measures a launch display against a baseline from a comparable earlier launch, since the exact product is new and has no sales history of its own. A factory outlet or Make to Stock shop like Delta Garments' own outlet, or Cedar & Co., measures against the same product's own recent sales, since it already has a track record. | A new product needs a comparable substitute baseline; an existing product can use its own direct history, which is a more reliable comparison. |
| Other changes at the same time | A price cut, a new season, a competitor's own promotion, or a delivery of new stock can all raise or lower sales at the same time as a display. | Any of these can be mistaken for the display's own effect if it is not tracked and separated out. |
| Display period length | A short display period, such as one week, is more exposed to a single unusual day; a longer period, such as four weeks, averages out day-to-day noise but delays the decision to change the display. | The period length is a trade-off between measurement reliability and how quickly a shop can act on the result. |
| Market | An export brand's own shops compare results across countries with different currencies and shopping patterns, so a lift is usually reported as a percentage change, not an absolute unit count. A single domestic shop can compare its own absolute unit counts directly. | Reporting method changes with scale, but the underlying baseline-versus-display comparison is the same. |
| Product family | A basic T-shirt's sell-through is easy to compare unit for unit; a multi-pack of socks or hosiery is usually measured by pack, not by individual pair; a bedding set is measured by set, since its pieces are not sold separately. | The counting unit changes by product family, but the method, comparing a display period to its own baseline, stays the same. |
Delta Garments runs a small factory outlet store next to its own site, selling second-quality pieces from orders such as Order A that do not meet Northline Apparel's own export standard but are still wearable. The outlet's merchandiser builds a table display of Order A's second-quality T-shirts, grouped by colour, and leaves it up for a fixed two-week display period.
No other change happens at the outlet during those two weeks: no price change, no new stock delivery, no season change. This makes the two weeks immediately before the display a fair baseline, since the only real difference between the two periods is the display itself.
The worked numberAn 86-piece lift over a clean, unchanged baseline is a real, measurable effect the display itself produced, since nothing else at the outlet changed between the two periods. If a price cut or new stock delivery had also happened during the display period, the same 86-piece rise could not be credited to the display alone without separating out the other cause's own share of the increase.
A display that sells 266 pieces sounds successful on its own, but without knowing that 180 of those pieces would likely have sold anyway, the shop cannot tell whether the display added real value or simply took credit for sales that were going to happen regardless. The cost of this mistake is not the display itself; it is that a shop keeps rebuilding displays that never actually lifted sales, while genuinely effective displays go unrecognized because no comparison was ever made.
Looking good is not the same as selling more. To know whether a display actually worked, compare what sold while it was up to what was already selling just before it went up. The difference between the two, not the display period's total sales on its own, is the display's real effect. If something else also changed at the same time, such as a price cut, the comparison is less reliable, and that other cause has to be considered before crediting the display.
The display of Order A's second-quality T-shirts shows the same 266 pieces sold over the two-week display period against a 180-piece baseline, an apparent lift of 86 pieces. Partway through the display, on day 3, the outlet also cut the second-quality T-shirt's price by 15% to clear ageing stock ahead of a new delivery.
Should the merchandiser report the full 86-piece lift as the display's own effect? Decide before opening the answer.
No. With a price cut running for most of the display period, the 86-piece lift cannot be credited to the display alone; the merchandiser should report it as a combined effect and flag the price cut separately.
1. The price cut ran for 11 of 14 display days, covering nearly all of the measured period, so its own effect on sales cannot be ignored or assumed to be small. 2. No price cut happened in the baseline period, so the baseline itself is still a fair starting point, but the display period is no longer a clean comparison once the price also changed. 3. Reporting the full lift as the display's own effect would overstate what the display achieved, and could lead the outlet to repeat a display design that had little effect of its own, while under-crediting the price cut that likely drove much of the increase.
The method's own step of checking for a confounding factor during the display period exists exactly for this situation; a lift that coincides with another known change should be reported as mixed, not attributed entirely to the display.
Order A's T-shirt carries a shop price of USD 19.90 per piece (docs/02 section 7). If Delta Garments' outlet wrongly credits its display design for the full 86-piece lift and rolls the same design out to other outlet tables expecting the same result without a matching price cut, and the true display-only lift is closer to 20 pieces, the shortfall per table becomes:
Repeating this mistake across even three more outlet tables, each expecting a lift that will not happen without the price cut, risks 198 pieces of overstocked, undersold second-quality product sitting unsold, a cost measured in tied-up stock and wasted display effort rather than in Order A's own export margin, but a real cost all the same.
Use the T22-C04 Display-performance tracker workbook, "Display sales measurement" sheet. For each display, record the baseline period sales, the display period sales, any confounding factor noted, and the calculated lift, so the whole outlet team can see which displays are proven and which are still guesses. Start with Delta Garments' own outlet example above, then add your own displays.
[Download the Display-performance tracker workbook](../../templates/T22-C04-display-performance-tracker.xlsx)
1. Why is a display's total sales figure not, by itself, a measure of its own effect?
2. A shop's baseline period showed 400 units sold in two weeks, and the display period showed 460 units sold in the following two weeks, with no other change at the shop. What is the display lift, expressed as a percentage?
3. A different shop's display period coincided with a new sales season, which usually raises sales there by about 10% on its own. How should this be handled?
| EN | AR | FR | TR | Meaning |
|---|---|---|---|---|
| Baseline sales | مبيعات خط الأساس | Ventes de référence | Referans satışlar | The sales a product would likely have made without the display, usually measured from a period just before it went up. |
| Sell-through rate | معدل البيع الفعلي | Taux d'écoulement | Satış oranı | The share of stock placed for a display that has actually sold by a given date. |
| Display lift | ارتفاع مبيعات العرض | Hausse due à la vitrine | Vitrin satış artışı | The extra sales a display produces above the baseline, credited to the display itself. |
| Attribution risk | خطر النسب الخاطئ | Risque d'attribution | Atfetme riski | The chance that a sales change is wrongly credited to a display when another cause, such as a price change, actually produced it. |
| Display period | فترة عرض العرض | Période d'exposition | Vitrin süresi | The fixed number of days or weeks a display stays up before it is measured or changed. |
| Confounding factor | عامل مربك | Facteur confondant | Karıştırıcı etken | Any other change happening at the same time as a display that could also explain a change in sales. |
Next you will learn how to use a display's own measured results to decide whether to keep it running as it is, change it, or take it down. Back to the course map