# NEW AND LOST VARIANCE

"New" variance is tied to combinations that have sales only in the second period. "Lost" variance is tied to combinations that have sales only in the first period. You can choose this option with the "show variance as" widget.

#### "LOST" AND "NEW" DEFINITION

Example. A combination might be the following:

Product: Shirts, Color: Red, Brand: Armani, City: Milan.

If this combination exists in the first year but this precise combination does not exist in the second year, the system will return a "Lost" variance value. Similarly if this exact combination does not exist in the first year but exists in the second, the system will return a "New" variance value. Changed volume variance refer to combinations that can be found with positive sales in both periods.

In case of "New" many things might have happen:

- We introduced the "shirts" product line
- We introduced "red" color in the "shirts" product line
- We started selling "Armani" brand apparel
- We opened a shop in Milan
- All or any combination of the above.

In other words, "Lost" and "New" simply mean that "something changed", but does not directly tell you anything about "what" exactly changed. It gives some indication of "stability". Please let us know if you find it useful.

#### "LOST" AND "NEW" VS PRICE VARIANCE

The app calculates variance "bottom" up, summing up the variance values of all combinations. If a combinations is "Lost" or "New" for that combination it is impossible to calculate the price component of the variance, because for one period units are zero and there is no price. Therefore for all Lost and New combinations, all the variance is booked as volume variance. This means that the price variance value calculated with this approach will generally be different from the price variance value calculated simply summing up all the rows and then computing the variance.