What is price elasticity and why lowering price does not always bring volume
Price elasticity is how much the quantity sold changes when the price changes, in percentage terms. It is not the same as survey-perceived sensitivity. Elasticity is measured on real sales, not on what the customer says they would do.
An elastic item loses a lot of sales for a small increase. An inelastic item barely reacts. The difference decides whether lowering the price brings volume or just gives margin away.
How the operation measures it
You compare the change in sales against the change in price, controlling for what gets in the way most: seasonality, the neighboring shelf’s promotion, the competitor’s move. Without that control, you measure noise and call it elasticity.
The number comes out per SKU, per store, per period. And here is the first trap: category elasticity is an abstraction no single item follows.
What the number hides
Average elasticity blends items that respond in opposite ways. Within the same category, one SKU can gain volume from a discount while its neighbor does not react at all. The average hides both.
It also hides the difference between promotional and base-price elasticity. The customer reacts strongly to a temporary discount and barely notices a shelf-price change. Treating the two as the same number is the mistake that makes margin drop with no volume showing up.
Why this changes an AI project
A pricing engine does not work with category elasticity. It works with elasticity per SKU and store, or it errs on the direction of the discount.
The acceptance criterion is not “the model estimated elasticity”. It is “margin rose without losing the volume that matters”, measured on sales. It is the same difference as real KVI versus a political list and planned versus forced markdown.
Think about the term each area of your operation uses but measures differently. “Price sensitivity” is usually that one. Commercial thinks it knows, marketing thinks it knows, and neither measures it on sales.
Tell me yours. In one hour I will send back a one-page card: the formula, the measurement points, what the number actually means, and where it misleads. If the card shows the way you estimate elasticity has to be rebuilt, the two-week Diagnóstico is the next step.