Related: Settings › Dimensions · Covariance · What If


Purpose

Some segments exist on one side of the comparison only: a country opened, a store closed, a product launched or discontinued. Left in the analysis, they distort every step — their whole weight lands in volume, then in mix and performance effects that mean nothing (a discontinued product looks like a price drop).

Out of scope segments takes them out of the steps and shows their contribution as one bar right after Start, the way controlling teams show a perimeter or scope effect before the like-for-like analysis:

Start  →  Out of scope (New + Discontinued)  →  [Comparable Start]  →  Volume  →  Price  → …  →  End

The market equation steps, mix / performance and covariance are then computed on the comparable (in-scope) rows only. Start and End keep the whole population, so the waterfall still adds up:

In-scope gap + Out of scope = Total gap


Turn it on

In Settings › Dimensions › Out of scope segments:

Option Description
Eligible dimensions Dimensions used to detect segments present on one side only. Leave empty to disable (default) — nothing is computed and your analysis is unchanged.
Scope label Rename the bar (default Out of scope, e.g. Perimeter effect or Store openings / closures). Shown once a dimension is selected.
Add comparable pillar On by default: adds a Comparable {Start} pillar right after the block, so you read the like-for-like starting point.

Click Apply to recompute.


How segments are classified

A segment is a combination of the values of the eligible dimensions (e.g. Country × Store).

Class Rule
New Every metric is absent or 0 on the Start side
Discontinued Every metric is absent or 0 on the End side
In scope Present on both sides — even with 1 unit at Start and 1 000 at End (the change then shows as volume / mix)

The eligible dimensions define the perimeter. With Country eligible, a country opening is out of scope; with only Zone eligible, the same opening stays inside its zone (the zone exists on both sides), so no block appears.

Detection runs after your dashboard filters: a segment filtered out on one side counts as absent there.


How to read it

  • Out of scope bar: net contribution of the perimeter change. Click it to open New segments and Discontinued segments, then click again to list the segments themselves.
  • Comparable pillar: Start restated on the comparable perimeter.
  • Steps: explain the like-for-like change only.
  • Smart comment: a paragraph right after the introduction describes the block (weight, number of new / discontinued segments, main contributors) when it is significant.
  • Dimensions then Steps analysis: the block is a root bar, and out-of-scope segments are removed from the other segments, so both analysis methods tell the same story.

Example

Italy opens, Canada closes, no price changes, eligible dimension = Country:

  Without the block With the block
Out of scope −23 500 (New IT +4 500, Discontinued CA −28 000)
Volume +4 000 +12 133
Average selling value −25 048 −8 865
Country performance (price) −12 844 0
Total gap −22 300 −22 300

Without the block, the closure of Canada shows up as a false price drop of −12 844. With the block, price performance is correctly 0.


Tips

  • Choose the finest dimension at which openings / closures happen (store, SKU, country) — not a grouping level.
  • Combine with Cascade mix effects along the hierarchy (Settings › Dimensions): the cascade then runs on the comparable perimeter.
  • There is no materiality threshold: a segment is out of scope only when it is strictly absent on one side.