Use the signed convention explicitly
Own-price elasticity describes the proportional quantity response to a proportional price change. For a conventional downward-sloping demand scenario, use a negative value. Some business reports show only its magnitude; a reported magnitude of 2 corresponds to −2 in MarginTrial's scenario calculator.
The constant-elasticity scenario is q1 = q0 × (p1 / p0)^ε. It keeps elasticity fixed throughout the interval, rather than applying a first-order percentage approximation to an arbitrarily large price change. That fixed elasticity is still an assumption, particularly outside observed price support.
Worked example: elastic revenue, lower contribution
A product sells 1,000 units at €50 with €25 variable cost. Assume elasticity −2 and reduce price to €45. The implied quantity is 1,234.57 units. Revenue rises, but the smaller contribution per unit leaves total contribution slightly below the original level.
| Measure | Original | Scenario |
|---|---|---|
| Price | €50 | €45 |
| Units | 1,000 | 1,234.57 |
| Revenue | €50,000 | €55,555.56 |
| Contribution | €25,000 | €24,691.36 |
| Contribution change | −€308.64 |
Ask what produced the price variation
A regression can associate lower prices with higher quantities because discounts coincide with advertising, better display or seasonal demand. Price can also respond to anticipated demand. Product and time controls are useful modeling choices, but their presence alone does not resolve every source of endogeneity.
Check whether meaningful price variation remains after the controls. A large coefficient supported by almost no conditional price variation is fragile. Review stock availability, influential observations, extrapolation and whether the same price measure was used consistently.
Design a price test around the decision
A randomized price or promotion experiment can provide a stronger design when feasible. Define the assignment unit, outcome and horizon before launch. If stores are assigned, store-level dependence matters; thousands of transactions within a few assigned stores are not thousands of independent assignments.
Interference can also matter when customers see different offers or stock moves between groups. Balance operational constraints with the intended estimand, and state what customer or market conditions the result can cover.
- Define the price or offer actually being varied.
- Choose contribution as well as revenue outcomes when relevant.
- Record eligibility, availability and concurrent campaigns.
- Analyze at the assignment level and retain the design record.
Map a scenario worksheet
Use the elasticity-price-cut synthetic case only as an assumed-volume illustration. The workbench accepts the resulting promo_units; it does not estimate elasticity from that row. Keep the elasticity and its evidence in the analyst note so later readers can reconstruct why volume changed.
| Assumption | Where to record | Example |
|---|---|---|
| Current price and volume | regular_price / baseline_units | 50 / 1000 |
| Scenario price | promo_price | 45 |
| Implied scenario quantity | promo_units | 1234.567901 |
| Signed elasticity and evidence | Report note / source analysis | −2; hypothetical assumption |
Frequently asked questions
Does elasticity below −1 mean a discount improves contribution? No. That classification concerns revenue response under the demand model. Contribution also depends on variable cost and campaign effects.
Can one elasticity be used at every price? Only as a stated extrapolation assumption. A local estimate need not describe distant prices, new customer segments or a different promotion mechanic.
Put it to work
Source context
Guide text and examples were prepared independently for MarginTrial. Sources provide the stated methodological or product context. They do not validate the hypothetical commercial outcomes.