Keyboard shortcuts

Press or to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Elasticity

A product's price elasticity of demand measures how strongly customer demand reacts when its price changes.

How to interpret Elasticity values

Why is elasticity always negative?

Because demand and price move in opposite directions: when price goes up, demand goes down; when price goes down, demand goes up. The model's demand curve is monotonically decreasing, meaning higher prices never predict higher sales volume.

Rule of thumb

Elasticity relates percentage price changes to percentage demand changes:

% Change in Demand ≈ Elasticity × % Change in Price

  • Example: An elasticity of -2.5 means that a 10% price decrease is predicted to increase demand by approximately +25% (-2.5 × -10% = +25%). Conversely, a 10% price increase would reduce demand by approximately -25% (-2.5 × +10% = -25%).

Interpretation guide

Elasticity RangePrice Sensitivity
-1.0 to -1.5Low sensitivity
-1.5 to -3.0Moderate sensitivity
-3.0 to -5.0High sensitivity
below -5.0Very high sensitivity

When is the value missing / not displayed?

If an optimization proposes keeping the current price unchanged (i.e. price change = 0), the elasticity cannot be evaluated using the transition formula and will be omitted.


Mathematical Definition

There are multiple definitions for measuring elasticity.

Example:

P r i c e 1 = P 1 = 100   E U R P r i c e 2 = P 2 = 120   E U R D e m a n d 1 = D 1 = 34   S a l e s D e m a n d 2 = D 2 = 26   S a l e s \begin{align} Demand_1 &= D_1 = 34\ Sales \\\\ Demand_2 &= D_2 = 26\ Sales \\\\ Price_1 &= P_1 = 100\ EUR \\\\ Price_2 &= P_2 = 120\ EUR \\\\ \end{align} S i m p l e E l a s t i c i t y = ( D 2 D 1 ) / D 1 ( P 2 P 1 ) / P 1 = ( 26 34 ) / 34 ( 120 100 ) / 100 = 1.18 E l a s t i c i t y = ( D 2 D 1 ) / D 1 + D 2 2 ( P 2 P 1 ) / P 1 + P 2 2 = ( 26 34 ) / 34 + 26 2 ( 120 100 ) / 100 + 120 2 = 1.47 \begin{align} Simple Elasticity = \frac{(D_2-D_1) / D_1}{(P_2-P_1) / P_1} &= \frac{(26-34) / 34}{(120-100) / 100} &= -1.18 \\\\ Elasticity = \frac{(D_2-D_1) / \frac{D_1+D_2}{2}}{(P_2-P_1) / \frac{P_1+P_2}{2}} &= \frac{(26-34) / \frac{34+26}{2}}{(120-100) / \frac{100+120}{2}} &= -1.47 \\\\ \end{align}

Our Elasticity KPI is using the second elasticity definition, which is often referred to as "Mid Point Elasticity". The advantage of this definition is that going back from P2 to P1 will give you the same elasticity, so it is symmetric. Also, the Simple Elasticity often comes up with very large elasticity outliers which are difficult to interpret.

We learn Elasticity on market, channel and product level based on product specific price change history and based on product attributes. If we don't have any information on the product specific price history, we approximate it using similar products. To ensure that we capture the true relationship between price changes and sales, we automatically detect seasonal patterns - like summer or holiday trends - and correct the data accordingly. For example, the sales of a seasonal product like sandals, which sells well in summer but less in winter, are smoothed out to account for their summer peak before calculating elasticity.